New energy station fault diagnosis system and method based on intelligent sensor

Abstract

The invention relates to the technical field of new energy station fault diagnosis, in particular to a new energy station fault diagnosis system and method based on an intelligent sensor, and the system comprises an intelligent processing unit which collects operation parameters in a station in real time, carries out the fault diagnosis of the operation parameters, and obtains diagnosis data; when the method is used, potential problems of equipment can be found in time, the data transmission bandwidth is effectively reduced, rapid transmission of real-time data is guaranteed, the equipment can adjust the working state in a self-adaptive mode under different operation conditions, false alarm or missing alarm caused by a fixed threshold value is avoided, and the service life of the equipment is prolonged. The influence of a network bottleneck on the overall efficiency of the system is reduced, the response speed and stability of the system are improved, the bandwidth resource waste is reduced, the overall performance of the network is improved, the timeliness and reliability of fault diagnosis are enhanced, meanwhile, the real-time performance and reliability of a data acquisition and monitoring system in a new energy station are greatly improved, and the application prospect is wide. And the potential cause of the fault can be accurately identified.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis of new energy stations, and in particular to a fault diagnosis system and method for new energy stations based on intelligent sensors. Background Art

[0002] New energy sites, especially photovoltaic power plants and wind farms, are easily affected by external factors due to the wide distribution of equipment and complex environment. Therefore, corresponding fault diagnosis systems are needed to quickly identify and locate faults through real-time monitoring, data analysis and intelligent algorithms, improve operation and maintenance efficiency, and ensure the stable operation of new energy sites.

[0003] The patent application number is CN202211469446.X, which states in the specification that "The embodiment of the present invention provides a fault handling method and system for a photovoltaic station, which belongs to the field of photovoltaic station operation and maintenance technology. The method includes: collecting the operation information of each equipment unit in the photovoltaic station, synchronizing the operation information of each equipment unit, and obtaining monitoring information; processing the monitoring information to determine whether there is an abnormality in the monitoring information; when there is an abnormality in the monitoring information, locating similar faults based on a preset secondary fault library, and determining and outputting a corresponding abnormal fault handling plan based on the located similar faults; if there is no similar fault location result, outputting an alarm information; in response to the trigger signal after the fault is eliminated, the alarm is stopped, the fault handling plan is recovered, and the secondary fault library is updated based on the fault handling plan. The scheme of the present invention improves the fault diagnosis and processing efficiency of photovoltaic stations, and also ensures the fault diagnosis accuracy of photovoltaic stations. "During the data collection and transmission process, the above technology will have delays or losses due to the distribution distance of equipment and changes in the outdoor environment. In particular, the data synchronization of long-distance equipment will be affected by network conditions or equipment conditions. In addition, the above technology relies on the professional level of operation and maintenance personnel, a large amount of data analysis and historical fault experience, resulting in the inability to systematically and intelligently replace manual experience and handle new fault types.

[0004] In summary, the development of a new energy station fault diagnosis system and method based on intelligent sensors is still a key issue that needs to be urgently solved in the field of new energy station fault diagnosis technology. Summary of the invention

[0005] The purpose of the present invention is to solve the problem in the prior art that during data collection and transmission, there will be delays or losses due to the distribution distance of equipment and changes in the outdoor environment, especially the data synchronization of long-distance equipment will be affected by network conditions or equipment conditions. In addition, the above-mentioned technology relies on the professional level of operation and maintenance personnel and a large amount of data analysis and historical failure experience, resulting in the inability to systematically and intelligently replace manual experience and handle new types of failures. The present invention provides a new energy station fault diagnosis system and method based on intelligent sensors.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The present invention provides a new energy station fault diagnosis system based on intelligent sensors, comprising: an intelligent processing unit, which collects operating parameters in the station in real time, and performs fault diagnosis on the operating parameters to obtain diagnostic data;

[0008] A network learning unit dynamically adjusts the transmission path of the operating parameters and the diagnostic data, and analyzes the diagnostic data;

[0009] A data inference unit, which fuses the operating parameters to obtain fused data, and performs multi-dimensional analysis on the fused data to obtain analysis data;

[0010] The decision support unit automatically generates operation and maintenance plans and makes system adjustments based on the analysis data.

[0011] Furthermore, the intelligent processing unit includes:

[0012] Intelligent collection module, through the deployment of various types of intelligent sensors, collects the operating parameters of each device in the station in real time;

[0013] A node processing module is provided at the operating equipment in the new energy station, and is used to perform real-time fault diagnosis on the operating parameters to obtain diagnostic data;

[0014] The eLearning modules include:

[0015] An automatic adjustment module dynamically adjusts the data transmission path and bandwidth according to the current network status and data transmission requirements, and selects a data transmission path that meets the data transmission requirements to transmit the operating parameters and diagnostic data;

[0016] The diagnosis module extracts important feature data and classifies them to obtain classified data by analyzing the diagnosis data.

[0017] Furthermore, the data reasoning unit includes:

[0018] A fusion analysis module is used to fuse the operating parameters from different sensors with the classification data, and then conduct multi-dimensional fault analysis, accurately locate the fault and comprehensively consider multiple factors to obtain analysis data;

[0019] A scenario simulation module creates various fault scenarios based on the analysis data and environmental data, simulates the response of the device under different fault scenarios, and thus forms a response plan for unknown and rare fault types;

[0020] The decision support unit includes:

[0021] A centralized monitoring module is used to centrally gather and compare the operating parameters and the analysis data to obtain comparison data, and predict the future operating state of the equipment according to the comparison data to obtain prediction data;

[0022] The system execution module identifies potential faults based on the prediction data, and provides and executes an operation and maintenance plan in combination with the response plan.

[0023] Furthermore, the workflow of the intelligent processing unit is:

[0024] The intelligent acquisition module deploys various types of sensors at key parts of the equipment to collect the operating parameters of each device, pre-process and compress the operating parameters according to real-time performance, and transmit the operating parameters to the node processing module wirelessly or wiredly. The pre-processing formula is: in is the filtered signal, w(j) is the filter coefficient, q(sj) is the original signal, T is the length of the filter, j is the discrete time index, and the compression formula is: Where E[q(s)] is the compressed data, x rj are the wavelet coefficients, is the mother wavelet function, R is a fixed maximum scale value, and J is the maximum number of translations allowed during transformation;

[0025] The node processing module performs preliminary calculation and analysis on the operating parameters obtained in real time, and performs data conversion, statistical analysis and trend analysis. It dynamically adjusts the device threshold based on the device specification and real-time feedback, and uses the standard deviation judgment formula to identify the fault mode in the local area of ​​the device according to the reasonable device threshold, and diagnoses and classifies the faults. The threshold dynamic adjustment formula is: S new =β·S old +(1-β)·S real , where S new is the new dynamic threshold, S old is the historical threshold, S real is the real-time monitoring value, β is the weighting factor, and the standard deviation determination formula is: Where χ is the mean, α is the standard deviation, and q j is the data point, I is the number of data points, and the fault classification formula is: y(q)=m S q+p, where y(q) is the classification result, m S is the support vector weight, q is the input feature, and p is the bias.

[0026] Furthermore, the workflow of the online learning unit is:

[0027] The automatic adjustment module continuously monitors the network status in real time, identifies network bottlenecks and load balancing problems, analyzes the real-time demand for data transmission, and automatically selects the data transmission path that meets the data transmission demand based on the current transmission demand and network load. It adopts a dynamic bandwidth allocation strategy to ensure the priority transmission of the operating parameters and diagnostic data. The network delay calculation formula is: Among them C network is the network delay, Total Delay is the total delay, TotalData is the total data volume, and the dynamic bandwidth allocation formula is: Where K new (s) is the new bandwidth requirement, K max is the maximum bandwidth, D req (s) is the real-time data demand, D total (s) is the total data volume;

[0028] The diagnosis module extracts important features from the diagnosis data as input data, performs time series analysis on the data, generates feature vectors to identify complex fault modes, and provides specific fault locations and impact ranges, and extracts important feature formulas:

[0029] Where Q(y) is the frequency domain representation of the signal, x(b) is the time domain signal, and l -i2πyb is a complex exponential function, i is the imaginary unit It is used to represent the frequency in complex form. 2πyb is the product of frequency y and time b. y is the frequency variable, b is the time domain variable, and pb is a small increment to the time variable b. The timing analysis formula is: Where q(s) is the observed value of the time series at time s, It represents the contribution of the past z values ​​of the time series to the current moment, z is the order of the model, γ j is the autoregressive coefficient, q(sj) represents the value of the time series at the past time sj, ∈(s) is the random error term at the current time s, and the complex fault identification formula is: where g s is the hidden state at the current moment, M g and N g is the weight matrix, qs is the current input data, p g For bias.

[0030] Furthermore, the workflow of the data reasoning unit is:

[0031] The fusion analysis module obtains the operating parameters and classification data from different sensors and environmental data, fuses the operating parameters and classification data into a unified data set, performs time alignment on the data set, processes the timeliness difference of the data, identifies the potential cause of the fault based on the data set, locates the specific fault point of the equipment with the help of the diagnosis model, and outputs a detailed fault analysis report. The data fusion formula is: in is the state prediction value, L is the state transfer matrix, F s+1s is the error covariance matrix, P is the process noise covariance, and the data time alignment formula is: Where q(s) is the aligned data, q0 and q1 are known data points, s1 and s0 are corresponding time points, and the fault location formula is: J = Y·Z, where J is the data after dimensionality reduction, Y is the original data matrix, and Z is the principal component matrix;

[0032] The scenario simulation module inputs the analysis data and environmental data, constructs multiple possible fault scenarios, simulates different fault types and their impact on equipment and systems, analyzes the response of equipment under various extreme conditions, and generates detailed operation manuals or automated response codes based on the simulation results. The fault scenario construction formula is: Where y(r) is the fault function, r j is the simulation input, B is the number of simulations, and the fault response simulation formula is: x(s) = t1·x·(s-1)+t2·x·(s-2)+…+v1·r·(s-1)+v2·r·(s-2)+∈(s), where x(s) is the system response, r(s) is the input signal, t i , v i is the model coefficient, ∈(s) is the noise term.

[0033] Furthermore, the workflow of the decision support unit is:

[0034] The centralized monitoring module standardizes and formats the operating parameters and analysis data from different sources, aggregates them into a central database for comparison, and then obtains comparative data. The prediction data of the equipment is generated based on the comparative data. The data standardization formula is: Where b is the standardized data, y is the original data, η is the mean, τ is the standard deviation, and the future state prediction formula is: E s =τ·(M E ·r s +Q E Es-1 +a E ), where E s is the hidden state at the current moment, M E and Q E is the weight matrix, r s is the input data, a E is bias;

[0035] The system execution module generates a specific program operation and maintenance strategy and then performs automated operations based on the response plan provided by the scenario simulation module and the prediction data provided by the centralized monitoring module.

[0036] On the other hand, the present invention also provides a new energy station fault diagnosis method based on intelligent sensors, which comprises the following steps:

[0037] S1. Real-time collection of operating parameters in the station, and fault diagnosis of the operating parameters to obtain diagnostic data;

[0038] S2. Dynamically adjust the transmission path of the operating parameters and diagnostic data, and analyze the diagnostic data;

[0039] S3, fusing the operating parameters to obtain fused data, and performing multi-dimensional analysis on the fused data to obtain analysis data;

[0040] S4. Automatically generate operation and maintenance plans and make system adjustments based on the analysis data.

[0041] Furthermore, in step S1, the operating parameters in the station are collected in real time, and fault diagnosis is performed on the operating parameters to obtain the diagnostic data in the following manner:

[0042] By deploying various types of intelligent sensors, the operating parameters of each device in the station can be collected in real time;

[0043] It is arranged at the operating equipment in the new energy station, and is used to perform real-time fault diagnosis on the operating parameters to obtain diagnostic data;

[0044] In step S2, the transmission path of the operating parameters and the diagnostic data is dynamically adjusted, and the method for analyzing the diagnostic data is:

[0045] According to the current network status and data transmission requirements, dynamically adjust the data transmission path and bandwidth, and select a data transmission path that meets the data transmission requirements to transmit the operating parameters and diagnostic data;

[0046] By analyzing the diagnostic data, important feature data is extracted and classified to obtain classified data.

[0047] Further, in step S3, the operating parameters are fused to obtain fused data, and a method for performing multi-dimensional analysis on the fused data to obtain analysis data is as follows:

[0048] It is used to integrate the operating parameters from different sensors with the classification data, and then conduct multi-dimensional fault analysis, accurately locate the fault and comprehensively consider multiple factors to obtain analysis data;

[0049] Creating various fault scenarios based on the analysis data and environmental data, simulating the response of the device under different fault scenarios, and thus forming response plans for unknown and rare fault types;

[0050] In step S4, the method of automatically generating an operation and maintenance plan and adjusting the system according to the analysis data is as follows:

[0051] Used to collect and compare the operating parameters and the analysis data to obtain comparison data, and predict the future operating state of the equipment according to the comparison data to obtain prediction data;

[0052] Potential failures are identified based on the prediction data, and an operation and maintenance plan is provided and executed in combination with the response plan.

[0053] Beneficial Effects

[0054] Compared with the known public technology, the technical solution provided by the present invention has the following beneficial effects:

[0055] When in use, the present invention is conducive to timely discovering potential problems of equipment, effectively reducing the data transmission bandwidth, ensuring the rapid transmission of real-time data, and enabling the equipment to adaptively adjust the working state under different operating conditions, avoiding false alarms or missed alarms due to fixed thresholds, and reducing the impact of network bottlenecks on the overall efficiency of the system, which is conducive to improving the response speed and stability of the system, reducing the waste of bandwidth resources, improving the overall performance of the network, and enhancing the timeliness and reliability of fault diagnosis. At the same time, it greatly improves the real-time and reliability of data acquisition and monitoring systems in new energy stations, which is conducive to accurately identifying potential causes of faults.

[0056] When in use, the present invention is conducive to greatly improving the accuracy and efficiency of fault diagnosis, providing response plans for operation and maintenance personnel in advance, improving the stability of the overall system, further improving the efficiency and accuracy of operation and maintenance work, and is conducive to identifying potential faults in advance, avoiding equipment downtime and production interruptions, and adjusting parameters in real time to ensure optimal operation results. Improve the efficiency and accuracy of operation and maintenance, and provide real-time feedback to adaptively optimize operation and maintenance strategies to ensure that the most appropriate operation plan is selected under different working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1This is a system diagram of a new energy station fault diagnosis system based on intelligent sensors of the present invention;

[0058] Figure 2 The present invention is a flow chart of a new energy station fault diagnosis method based on intelligent sensors. DETAILED DESCRIPTION

[0059] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0060] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0061] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0062] Embodiment 1:

[0063] like Figure 1 As shown, the present invention provides a new energy station fault diagnosis system based on intelligent sensors, including: an intelligent processing unit, which collects operating parameters in the station in real time, and performs fault diagnosis on the operating parameters to obtain diagnostic data;

[0064] The intelligent processing unit includes:

[0065] Intelligent collection module, through the deployment of various types of intelligent sensors, collects the operating parameters of each device in the station in real time;

[0066] A node processing module is provided at the operating equipment in the new energy station, and is used to perform real-time fault diagnosis on the operating parameters to obtain diagnostic data;

[0067] The workflow of the intelligent processing unit is:

[0068] The intelligent acquisition module deploys various types of sensors at key parts of the equipment to collect the operating parameters of each device, pre-process and compress the operating parameters according to real-time performance, and transmit the operating parameters to the node processing module wirelessly or wiredly. The pre-processing formula is: in is the filtered signal, w(j) is the filter coefficient, q(sj) is the original signal, T is the length of the filter, j is the discrete time index, and the compression formula is: Where E[q(s)] is the compressed data, x rj are the wavelet coefficients, is the mother wavelet function, R is a fixed maximum scale value, and J is the maximum number of translations allowed during transformation;

[0069] The node processing module performs preliminary calculation and analysis on the operating parameters obtained in real time, and performs data conversion, statistical analysis and trend analysis. It dynamically adjusts the device threshold based on the device specification and real-time feedback, and uses the standard deviation judgment formula to identify the fault mode in the local area of ​​the device according to the reasonable device threshold, and diagnoses and classifies the faults. The threshold dynamic adjustment formula is: S new =β·S old +(1-β)·S real , where S new is the new dynamic threshold, S old is the historical threshold, S real is the real-time monitoring value, β is the weighting factor, and the standard deviation determination formula is: Where χ is the mean, α is the standard deviation, and q j is the data point, I is the number of data points, and the fault classification formula is: y(q)=m S q+p, where y(q) is the classification result, m S is the support vector weight, q is the input feature, and p is the bias;

[0070] Specifically, various types of intelligent sensors are deployed at key parts of the equipment through the intelligent acquisition module. These sensors can collect various operating parameters of the equipment in real time, including but not limited to operating data such as temperature, pressure, current, and vibration. The data acquisition cycle and frequency are set according to the operating characteristics of the equipment to ensure real-time and data integrity. At the same time, the original data is filtered and compressed, and the collected data is received and analyzed in real time through the node processing module. The collected data is preliminarily calculated, converted, statistically analyzed, and trended. The threshold of the equipment is dynamically adjusted according to the historical data, specifications, and real-time feedback of the equipment to ensure that the equipment operates within a healthy range, which is conducive to timely discovering potential problems of the equipment, effectively reducing the data transmission bandwidth, and ensuring the rapid transmission of real-time data. The equipment can adaptively adjust the working status under different operating conditions, avoid false alarms or omissions caused by fixed thresholds, and reduce the impact of network bottlenecks on the overall efficiency of the system.

[0071] A network learning unit dynamically adjusts the transmission path of the operating parameters and the diagnostic data, and analyzes the diagnostic data;

[0072] The eLearning modules include:

[0073] An automatic adjustment module dynamically adjusts the data transmission path and bandwidth according to the current network status and data transmission requirements, and selects a data transmission path that meets the data transmission requirements to transmit the operating parameters and diagnostic data;

[0074] The diagnosis module extracts important feature data and classifies them to obtain classified data by analyzing the diagnosis data;

[0075] The workflow of the e-learning unit is:

[0076] The automatic adjustment module continuously monitors the network status in real time, identifies network bottlenecks and load balancing problems, analyzes the real-time demand for data transmission, and automatically selects the data transmission path that meets the data transmission demand based on the current transmission demand and network load. It adopts a dynamic bandwidth allocation strategy to ensure the priority transmission of the operating parameters and diagnostic data. The network delay calculation formula is: Among them C network is the network delay, Total Delay is the total delay, TotalData is the total data volume, and the dynamic bandwidth allocation formula is: Where K new (s) is the new bandwidth requirement, K max is the maximum bandwidth, D req (s) is the real-time data demand, D total (s) is the total data volume;

[0077] The diagnosis module extracts important features from the diagnosis data as input data, performs time series analysis on the data, generates feature vectors to identify complex fault modes, and provides specific fault locations and impact ranges, and extracts important feature formulas:

[0078] Where Q(y) is the frequency domain representation of the signal, x(b) is the time domain signal, and l -i2πyb is a complex exponential function, i is the imaginary unit It is used to represent the frequency in complex form. 2πyb is the product of frequency y and time b. y is the frequency variable, b is the time domain variable, and pb is a small increment to the time variable b. The timing analysis formula is: Where q(s) is the observed value of the time series at time s, It represents the contribution of the past z values ​​of the time series to the current moment, z is the order of the model, γ j is the autoregressive coefficient, q(sj) represents the value of the time series at the past time sj, ∈(s) is the random error term at the current time s, and the complex fault identification formula is: where g s is the hidden state at the current moment, M g and N g is the weight matrix, q s is the current input data, p g is bias;

[0079] Specifically, the automatic adjustment module continuously monitors the current network status, including but not limited to bandwidth usage, data transmission delay, and network bottleneck, to ensure that data can be transmitted on the optimal path, and automatically adjusts the bandwidth according to real-time data requirements and current network load, and prioritizes the transmission of key data. It automatically selects the transmission path that meets the data transmission requirements to ensure that the data can reach the target device efficiently, and uses the diagnostic model to extract important features from the diagnostic data for further timing analysis and fault identification. The feature data is analyzed to identify complex fault modes and provide the fault location and impact range, which is conducive to improving the response speed and stability of the system, reducing the waste of bandwidth resources, improving the overall network performance, and enhancing the timeliness and reliability of fault diagnosis. At the same time, it greatly improves the real-time and reliability of the data acquisition and monitoring systems in new energy stations.

[0080] A data inference unit, which fuses the operating parameters to obtain fused data, and performs multi-dimensional analysis on the fused data to obtain analysis data;

[0081] The data reasoning unit includes:

[0082] A fusion analysis module is used to fuse the operating parameters from different sensors with the classification data, and then conduct multi-dimensional fault analysis, accurately locate the fault and comprehensively consider multiple factors to obtain analysis data;

[0083] A scenario simulation module creates various fault scenarios based on the analysis data and environmental data, simulates the response of the device under different fault scenarios, and thus forms a response plan for unknown and rare fault types;

[0084] The workflow of the data reasoning unit is:

[0085] The fusion analysis module obtains the operating parameters and classification data from different sensors and environmental data, fuses the operating parameters and classification data into a unified data set, performs time alignment on the data set, processes the timeliness difference of the data, identifies the potential cause of the fault based on the data set, locates the specific fault point of the equipment with the help of the diagnosis model, and outputs a detailed fault analysis report. The data fusion formula is: in is the state prediction value, L is the state transfer matrix, F s+1s is the error covariance matrix, P is the process noise covariance, and the data time alignment formula is: Where q(s) is the aligned data, q0 and q1 are known data points, s1 and s0 are corresponding time points, and the fault location formula is: J = Y·Z, where J is the data after dimensionality reduction, Y is the original data matrix, and Z is the principal component matrix;

[0086] The scenario simulation module inputs the analysis data and environmental data, constructs multiple possible fault scenarios, simulates different fault types and their impact on equipment and systems, analyzes the response of equipment under various extreme conditions, and generates detailed operation manuals or automated response codes based on the simulation results. The fault scenario construction formula is: Where y(r) is the fault function, r j is the simulation input, B is the number of simulations, and the fault response simulation formula is: x(s) = t1·x·(s-1)+t2·x·(s-2)+…+v1·r·(s-1)+v2·r·(s-2)+∈(s), where x(s) is the system response, r(s) is the input signal, t i , v i is the model coefficient, ∈(s) is the noise term;

[0087] Specifically, the data from multiple sensors is combined with environmental data through the fusion analysis module, and a unified data set is formed through time alignment processing. The cause of the fault is identified based on the fused data, and the specific fault point is located in conjunction with the diagnosis model to generate a detailed fault analysis report. The scenario simulation module is used to construct a variety of fault scenarios based on historical fault data, equipment parameters and real-time monitoring data, which is conducive to accurately identifying the potential causes of faults, greatly improving the accuracy and efficiency of fault diagnosis, providing response plans for operation and maintenance personnel in advance, improving the stability of the overall system, and further improving the efficiency and accuracy of operation and maintenance work.

[0088] The decision support unit automatically generates operation and maintenance plans and makes system adjustments based on the analysis data;

[0089] The decision support unit includes:

[0090] A centralized monitoring module is used to centrally gather and compare the operating parameters and the analysis data to obtain comparison data, and predict the future operating state of the equipment according to the comparison data to obtain prediction data;

[0091] A system execution module identifies potential faults based on the prediction data, and provides and executes an operation and maintenance plan in combination with the response plan;

[0092] The workflow of the decision support unit is:

[0093] The centralized monitoring module standardizes and formats the operating parameters and analysis data from different sources, aggregates them into a central database for comparison, and then obtains comparative data. The prediction data of the equipment is generated based on the comparative data. The data standardization formula is: Where b is the standardized data, y is the original data, η is the mean, τ is the standard deviation, and the future state prediction formula is: E s =τ·(M E ·r s +Q E E s-1 +a E ), where E s is the hidden state at the current moment, M E and Q E is the weight matrix, r s is the input data, a E is bias;

[0094] The system execution module generates a specific program operation and maintenance strategy and then performs automated operations based on the response plan provided by the scenario simulation module and the prediction data provided by the centralized monitoring module;

[0095] Specifically, the centralized monitoring module collects data from various sensors and performs standardized processing to eliminate differences between different devices, sensors, and data sources. Historical data is combined with real-time data to predict the future operating status of the equipment. Specific operation and maintenance strategies are generated based on the data provided by the centralized monitoring module and the scenario simulation module. During the execution process, the operation effect is monitored in real time, parameters are adjusted, and necessary supplementary measures are taken. Feedback data is collected, strategy verification is performed, and updates and optimizations are performed. This is conducive to identifying potential faults in advance, avoiding equipment downtime and production interruptions, adjusting parameters in real time, ensuring optimal operation effects, improving the efficiency and accuracy of operation and maintenance, and providing real-time feedback to adaptively optimize operation and maintenance strategies to ensure that the most appropriate operation plan is selected under different working conditions.

[0096] Embodiment 2:

[0097] like Figure 2 As shown, embodiment 2 provides a new energy station fault diagnosis method based on intelligent sensors, which includes the following steps:

[0098] S1. Real-time collection of operating parameters in the station, and fault diagnosis of the operating parameters to obtain diagnostic data;

[0099] S2. Dynamically adjust the transmission path of the operating parameters and diagnostic data, and analyze the diagnostic data;

[0100] S3, fusing the operating parameters to obtain fused data, and performing multi-dimensional analysis on the fused data to obtain analysis data;

[0101] S4. Automatically generate operation and maintenance plans and make system adjustments based on the analysis data.

[0102] Furthermore, in step S1, the operating parameters in the station are collected in real time, and fault diagnosis is performed on the operating parameters to obtain the diagnostic data in the following manner:

[0103] By deploying various types of intelligent sensors, the operating parameters of each device in the station can be collected in real time;

[0104] It is arranged at the operating equipment in the new energy station, and is used to perform real-time fault diagnosis on the operating parameters to obtain diagnostic data;

[0105] In step S2, the transmission path of the operating parameters and the diagnostic data is dynamically adjusted, and the method for analyzing the diagnostic data is:

[0106] According to the current network status and data transmission requirements, dynamically adjust the data transmission path and bandwidth, and select a data transmission path that meets the data transmission requirements to transmit the operating parameters and diagnostic data;

[0107] By analyzing the diagnostic data, important feature data is extracted and classified to obtain classified data.

[0108] Further, in step S3, the operating parameters are fused to obtain fused data, and a method for performing multi-dimensional analysis on the fused data to obtain analysis data is as follows:

[0109] It is used to integrate the operating parameters from different sensors with the classification data, and then conduct multi-dimensional fault analysis, accurately locate the fault and comprehensively consider multiple factors to obtain analysis data;

[0110] Creating various fault scenarios based on the analysis data and environmental data, simulating the response of the device under different fault scenarios, and thus forming response plans for unknown and rare fault types;

[0111] In step S4, the method of automatically generating an operation and maintenance plan and adjusting the system according to the analysis data is as follows:

[0112] Used to collect and compare the operating parameters and the analysis data to obtain comparison data, and predict the future operating state of the equipment according to the comparison data to obtain prediction data;

[0113] Potential failures are identified based on the prediction data, and an operation and maintenance plan is provided and executed in combination with the response plan.

[0114] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A new energy station fault diagnosis system based on intelligent sensors, characterized in that: include: An intelligent processing unit collects operating parameters in the station in real time, performs fault diagnosis on the operating parameters, and obtains diagnostic data; A network learning unit dynamically adjusts the transmission path of the operating parameters and the diagnostic data, and analyzes the diagnostic data; A data inference unit, which fuses the operating parameters to obtain fused data, and performs multi-dimensional analysis on the fused data to obtain analysis data; The decision support unit automatically generates operation and maintenance plans and makes system adjustments based on the analysis data.

2. According to claim 1, a new energy station fault diagnosis system based on intelligent sensors is characterized in that: The intelligent processing unit includes: Intelligent collection module, through the deployment of various types of intelligent sensors, collects the operating parameters of each device in the station in real time; A node processing module is provided at the operating equipment in the new energy station, and is used to perform real-time fault diagnosis on the operating parameters to obtain diagnostic data; The eLearning modules include: An automatic adjustment module dynamically adjusts the data transmission path and bandwidth according to the current network status and data transmission requirements, and selects a data transmission path that meets the data transmission requirements to transmit the operating parameters and diagnostic data; The diagnosis module extracts important feature data and classifies them to obtain classified data by analyzing the diagnosis data.

3. According to claim 2, a new energy station fault diagnosis system based on intelligent sensors is characterized in that: The data reasoning unit includes: A fusion analysis module is used to fuse the operating parameters from different sensors with the classification data, and then conduct multi-dimensional fault analysis, accurately locate the fault and comprehensively consider multiple factors to obtain analysis data; A scenario simulation module creates various fault scenarios based on the analysis data and environmental data, simulates the response of the device under different fault scenarios, and thus forms a response plan for unknown and rare fault types; The decision support unit includes: A centralized monitoring module is used to centrally gather and compare the operating parameters and the analysis data to obtain comparison data, and predict the future operating state of the equipment according to the comparison data to obtain prediction data; The system execution module identifies potential faults based on the prediction data, and provides and executes an operation and maintenance plan in combination with the response plan.

4. A new energy station fault diagnosis system based on intelligent sensors according to claim 3, characterized in that: The workflow of the intelligent processing unit is: The intelligent acquisition module deploys various types of sensors at key parts of the equipment to collect the operating parameters of each device, pre-process and compress the operating parameters according to real-time performance, and transmit the operating parameters to the node processing module wirelessly or wiredly. The pre-processing formula is: in is the filtered signal, w(j) is the filter coefficient, q(sj) is the original signal, T is the length of the filter, j is the discrete time index, and the compression formula is: Where E[q(s)] is the compressed data, x rj are the wavelet coefficients, is the mother wavelet function, R is a fixed maximum scale value, and J is the maximum number of translations allowed during transformation; The node processing module performs preliminary calculation and analysis on the operating parameters obtained in real time, and performs data conversion, statistical analysis and trend analysis. It dynamically adjusts the device threshold based on the device specification and real-time feedback, and uses the standard deviation judgment formula to identify the fault mode in the local area of ​​the device according to the reasonable device threshold, and diagnoses and classifies the faults. The threshold dynamic adjustment formula is: S new =β·S old +(1-β)·S real , where S new is the new dynamic threshold, S old is the historical threshold, S real is the real-time monitoring value, β is the weighting factor, and the standard deviation determination formula is: Where χ is the mean, α is the standard deviation, and q j is the data point, I is the number of data points, and the fault classification formula is: y(q)=m S q+p, where y(q) is the classification result, m S is the support vector weight, q is the input feature, and p is the bias.

5. A new energy station fault diagnosis system based on intelligent sensors according to claim 4, characterized in that: The workflow of the e-learning unit is: The automatic adjustment module continuously monitors the network status in real time, identifies network bottlenecks and load balancing problems, analyzes the real-time demand for data transmission, and automatically selects the data transmission path that meets the data transmission demand based on the current transmission demand and network load. It adopts a dynamic bandwidth allocation strategy to ensure the priority transmission of the operating parameters and diagnostic data. The network delay calculation formula is: Among them C network is the network delay, TotalDelay is the total delay, TotalData is the total data volume, and the dynamic bandwidth allocation formula is: Where K new (s) is the new bandwidth requirement, K max is the maximum bandwidth, D req (s) is the real-time data demand, D total (s) is the total data volume; The diagnosis module extracts important features from the diagnosis data as input data, performs time series analysis on the data, generates feature vectors to identify complex fault modes, and provides specific fault locations and impact ranges, and extracts important feature formulas: Where Q(y) is the frequency domain representation of the signal, x(b) is the time domain signal, and l -i2πyb is a complex exponential function, i is the imaginary unit It is used to represent the frequency in complex form. 2πyb is the product of frequency y and time b. y is the frequency variable, b is the time domain variable, and pb is a small increment to the time variable b. The timing analysis formula is: Where q(s) is the observed value of the time series at time s, It represents the contribution of the past z values ​​of the time series to the current moment, z is the order of the model, γ j is the autoregressive coefficient, q(sj) represents the value of the time series at the past time sj, ∈(s) is the random error term at the current time s, and the complex fault identification formula is: where g s is the hidden state at the current moment, M g and N g is the weight matrix, q s is the current input data, p g For bias.

6. A new energy station fault diagnosis system based on intelligent sensors according to claim 5, characterized in that: The workflow of the data reasoning unit is: The fusion analysis module obtains the operating parameters and classification data from different sensors and environmental data, fuses the operating parameters and classification data into a unified data set, performs time alignment on the data set, processes the timeliness difference of the data, identifies the potential cause of the fault based on the data set, locates the specific fault point of the equipment with the help of the diagnosis model, and outputs a detailed fault analysis report. The data fusion formula is: in is the state prediction value, L is the state transfer matrix, F s+1s is the error covariance matrix, P is the process noise covariance, and the data time alignment formula is: Where q(s) is the aligned data, q0 and q1 are known data points, s1 and s0 are corresponding time points, and the fault location formula is: J = Y·Z, where J is the data after dimensionality reduction, Y is the original data matrix, and Z is the principal component matrix; The scenario simulation module inputs the analysis data and environmental data, constructs multiple possible fault scenarios, simulates different fault types and their impact on equipment and systems, analyzes the response of equipment under various extreme conditions, and generates detailed operation manuals or automated response codes based on the simulation results. The fault scenario construction formula is: Where B is the number of samples in the dataset, r j is the jth data point in the data set, y(r) represents the output value of a function, It means that for the function y(r) at every r j The sum of the values ​​at the data points, y(r j ) means applying function y to the jth data point r j The results obtained above are as follows: the fault response simulation formula is: x(s) = t1·x·(s-1)+t2·x·(s-2)+…+v1·r·(s-1)+v2·r·(s-2)+∈(s), where x(s) is the target variable to be calculated at time s, x·(s-1), x·(s-2), … are the past values ​​of the target variable x(s), r·(s-1), r·(s-2), … are the external input variables affecting the target variable x(s), t1, t2 are coefficients related to the past outputs x·(s-1), x·(s-2), …, v1, v2 are coefficients related to the external inputs r·(s-1), r·(s-2), …, and ∈(s) is the noise term.

7. A new energy station fault diagnosis system based on intelligent sensors according to claim 6, characterized in that: The workflow of the decision support unit is: The centralized monitoring module standardizes and formats the operating parameters and analysis data from different sources, aggregates them into a central database for comparison, and then obtains comparative data. The prediction data of the equipment is generated based on the comparative data. The data standardization formula is: Where b is the standardized data, y is the original data, η is the mean, τ is the standard deviation, and the future state prediction formula is: E s =τ·(M E ·r s +Q E E s-1 +a E ), where E s is the hidden state at the current moment, M E and Q E is the weight matrix, r s is the input data, a E is bias; The system execution module generates a specific program operation and maintenance strategy and then performs automated operations based on the response plan provided by the scenario simulation module and the prediction data provided by the centralized monitoring module.

8. A fault diagnosis method for a new energy station based on an intelligent sensor, based on a fault diagnosis system for a new energy station based on an intelligent sensor according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1. Real-time collection of operating parameters in the station, and fault diagnosis of the operating parameters to obtain diagnostic data; S2. Dynamically adjust the transmission path of the operating parameters and diagnostic data, and analyze the diagnostic data; S3, fusing the operating parameters to obtain fused data, and performing multi-dimensional analysis on the fused data to obtain analysis data; S4. Automatically generate operation and maintenance plans and make system adjustments based on the analysis data.

9. A new energy station fault diagnosis method based on intelligent sensors according to claim 8, characterized in that: In step S1, the operating parameters in the station are collected in real time, and fault diagnosis is performed on the operating parameters to obtain diagnostic data in the following manner: By deploying various types of intelligent sensors, the operating parameters of each device in the station can be collected in real time; It is arranged at the operating equipment in the new energy station, and is used to perform real-time fault diagnosis on the operating parameters to obtain diagnostic data; In step S2, the transmission path of the operating parameters and the diagnostic data is dynamically adjusted, and the method for analyzing the diagnostic data is: According to the current network status and data transmission requirements, dynamically adjust the data transmission path and bandwidth, and select a data transmission path that meets the data transmission requirements to transmit the operating parameters and diagnostic data; By analyzing the diagnostic data, important feature data is extracted and classified to obtain classified data.

10. A fault diagnosis method for a new energy station based on an intelligent sensor according to claim 9, characterized in that: In step S3, the operating parameters are fused to obtain fused data, and a method for performing multi-dimensional analysis on the fused data to obtain analysis data is as follows: It is used to integrate the operating parameters from different sensors with the classification data, and then conduct multi-dimensional fault analysis, accurately locate the fault and comprehensively consider multiple factors to obtain analysis data; Creating various fault scenarios based on the analysis data and environmental data, simulating the response of the device under different fault scenarios, and thus forming response plans for unknown and rare fault types; In step S4, the method of automatically generating an operation and maintenance plan and adjusting the system according to the analysis data is as follows: Used to collect and compare the operating parameters and the analysis data to obtain comparison data, and predict the future operating state of the equipment according to the comparison data to obtain prediction data; Potential failures are identified based on the prediction data, and an operation and maintenance plan is provided and executed in combination with the response plan.

Images