New energy station environment monitoring system and method based on chemical sensor

Abstract

The invention relates to the technical field of new energy station environment monitoring, in particular to a new energy station environment monitoring system and method based on a chemical sensor, and the system comprises an edge calculation unit which carries out the fusion use of a plurality of sensors, and dynamically adjusts the data standardization parameters. The real-time performance is improved, different working conditions are adapted, the influence caused by repeated data is reduced, the abnormal judgment threshold value is dynamically adjusted according to real-time data fluctuation and seasonal change, the limitation of a traditional static threshold value is avoided, it is ensured that the precision of the sensor is always in the optimal state, and the accuracy of the sensor is improved. Errors caused by environmental changes are reduced, reliability and maintenance efficiency of the system are improved, sensor faults are recognized, it is guaranteed that even if individual sensors fail, the system can still provide reliable data support through correlation analysis, and energy consumption is reduced and power generation efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy station environment monitoring, and in particular to a new energy station environment monitoring system and method based on chemical sensors. Background Art

[0002] Chemical sensors are devices that can detect and respond to specific chemicals. They are widely used in environmental monitoring, industrial process control, medical diagnosis, food safety and other fields. Chemical sensors can be used to monitor and obtain real-time environmental parameters around new energy power generation facilities such as solar photovoltaic power stations, wind farms, and biomass power plants.

[0003] The patent application number is 201910803721.9, which states in the specification that “the present invention provides a remote monitoring system for a new energy source, comprising: a monitoring terminal, a network-side server and a user terminal; wherein the monitoring terminal is used to monitor the working environment of the new energy source, obtain working environment information, and transmit the working environment information to the network-side server; the network-side server is used to determine whether there is an abnormality in the working environment of the new energy source according to the working environment information transmitted by the monitoring terminal, obtain the abnormal information of the working environment of the new energy source, and transmit the acquired abnormal information and the working environment information transmitted by the monitoring terminal to the user terminal; the user terminal is used to display the abnormal information and the working environment information to the staff; thereby realizing the staff of the user terminal Remote monitoring of the working environment of new energy by personnel". Although the above system collects a variety of working environment information through the monitoring terminal and standardizes the data, in actual applications, the data of different sensors have differences in accuracy and reliability. At the same time, some important data details will be lost in the standardization process. Especially in extreme working environments, the measurement differences of different sensors will even affect the judgment results. In addition, the above system judges whether there is an abnormality by comparing the site environment data with the abnormal data in the database. The abnormality judgment depends on the preset threshold and historical data, which requires the abnormal database to contain enough sample data and set reasonable thresholds. Some newly established sites will face the situation of insufficient sample data, which will lead to false alarms or missed alarms.

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

[0005] The purpose of the present invention is to solve the problem in the prior art that although the above-mentioned system collects a variety of working environment information through the monitoring end and standardizes the data, in actual application, the data of different sensors have differences in accuracy and reliability. At the same time, some important data details may be lost during the standardization process. Especially in extreme working environments, the measurement differences of different sensors may even affect the judgment results. In addition, the above-mentioned system determines whether there is an abnormality by comparing the station environment data with the abnormal data in the database, and the abnormality judgment depends on preset thresholds and historical data. This requires that the abnormal database needs to contain enough sample data and set reasonable thresholds. Some newly established stations may face the situation of not having enough sample data, which may lead to false alarms or missed alarms. The present invention provides a new energy station environment monitoring system and method based on chemical sensors.

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

[0007] The present invention provides a new energy station environment monitoring system based on chemical sensors, including: an edge computing unit, which fuses multiple sensors for use, dynamically adjusts data standardization parameters, and transmits only feature data;

[0008] The cross-region access unit performs analysis based on the characteristic data to obtain abnormal sample data and dynamically adjust the abnormal determination threshold;

[0009] A self-monitoring unit that performs regular self-calibration according to the abnormality determination threshold, automatically diagnoses and generates a repair plan;

[0010] The intelligent collaboration unit makes collaborative anomaly determinations based on the characteristic data and intelligently optimizes the equipment.

[0011] Furthermore, the edge computing unit includes:

[0012] Dynamic integration system integrates multiple sensors and uses machine learning algorithms to dynamically adjust data standardization parameters based on sensor characteristics and environmental changes, eliminating sensor accuracy differences and obtaining standard data.

[0013] An edge mixing module is used to pre-process the standard data, transmit only data with key features, and add different communication modes to switch between different environments;

[0014] Cross-region access units include:

[0015] A distributed data platform is used to access and obtain the characteristic data of each new energy station in real time, integrate and process the characteristic data of each station, and realize seamless connection of data from different stations through standardized data formats and protocols;

[0016] The feature analysis training module analyzes, deeply mines and recognizes patterns of the acquired feature data, identifies the common features and individual features of the feature data, and further analyzes and trains the abnormality determination threshold.

[0017] Further, the self-monitoring unit comprises:

[0018] An automatic calibration module, which automatically calibrates the sensor regularly based on the real-time change of the abnormality determination threshold value and adapts the sensor to the environment;

[0019] A self-diagnosis module, which automatically diagnoses and reports the cause of the abnormal state based on the real-time operating state of the sensor after the automatic calibration, and generates a repair plan to perform software maintenance or hardware maintenance;

[0020] The intelligent collaborative unit includes:

[0021] The redundancy judgment module performs correlation analysis on the characteristic data of multiple sensors through data association analysis and fusion judgment, so as to understand the real environmental change reaction data, and trigger an alarm when the characteristic data of multiple sensors are abnormal;

[0022] An intelligent scheduling system intelligently schedules equipment according to the environmental change response data.

[0023] Furthermore, the workflow of the edge computing unit includes:

[0024] The dynamic integration system is used to process data from multiple sensors, integrating the data of temperature, humidity, air pressure and chemical concentration into a data pool. During the fusion process, the weight of each sensor is updated through the weighted fusion mechanism and the accuracy, stability and environmental changes of each sensor. The weighted average algorithm formula is: where q fin is the fused output, e j is the measurement value of the jth sensor, w j is the weight of the sensor, u is the number of sensors, and the weight update formula is: in is the noise variance of sensor j at time s;

[0025] The edge mixing module is used to analyze and process the standard data, use principal component analysis to reduce the dimension of the standard data, extract the most representative feature data, and perform outlier detection to determine whether the feature data exceeds a reasonable range, screen out the feature data with high correlation, reduce unnecessary data transmission, and use dynamic optimization formulas to reduce bandwidth requirements and improve system real-time performance. Under different environmental conditions, the communication mode switching formula will select the most appropriate communication protocol based on the current bandwidth requirements, delay requirements, and energy consumption constraints. The principal component analysis formula is: W pca =R S (W-β), where W pca is the data after dimensionality reduction, R is the principal component matrix, β is the mean of the data, and the anomaly detection formula is: Where T j is the T value of the jth data, is the observed value, β and α are the mean and standard deviation of the data set respectively, and the dynamic optimization formula is: U eff =U orig × cpn , where U eff is the effective transmission rate, U orig is the original data transmission rate, χ cpn For compression efficiency, the communication mode switching formula is: Minimize y(B, L, P) = m1×B+m2×L+m3×P, where m1, m2, m3 are weight factors used to adjust the relative importance of bandwidth, delay and power consumption, and B, L, P are network bandwidth, delay and power consumption.

[0026] Furthermore, the workflow of the cross-region access unit includes:

[0027] The distributed data platform collects the characteristic data by accessing the database of each station in real time. The characteristic data format and protocol are deduplicated and cleaned through the hash algorithm formula, so that the characteristic data of different stations can ensure data uniqueness. At the same time, in the process of transmitting the characteristic data, the characteristic data of each area is aggregated using the weighted average formula to ensure the accuracy and completeness of the characteristic data. The hash algorithm formula is: I(o)=hash(o), where I(o) is the result of hashing data o. The weighted average formula is: Where P agg is the aggregated data, d j is the data from the jth station, a j is the corresponding weight;

[0028] The feature analysis training module uses big data analysis to conduct in-depth mining of the massive feature data collected across regions, analyze the feature data of each station, identify and extract the common features and individual features of the feature data, and discover the potential laws of environmental changes and sensor failures between stations, so as to dynamically adjust the abnormal judgment threshold. The formula for deep data mining is: Where B is the number of clusters, D b is the number of samples in the b-th cluster, δb is the center of the b-th cluster, and the dynamic anomaly threshold adjustment formula is: ε asd =ε base ×(1+φ), where ε asd is the adjusted threshold, ε base is the initial threshold, and φ is the dynamic adjustment factor calculated according to data fluctuations.

[0029] Furthermore, the workflow of the self-monitoring unit includes:

[0030] The automatic calibration module performs automatic calibration according to the preset cycle, real-time feedback, and real-time changes in the abnormal judgment threshold. In the process of automatic calibration, the least square method is used to optimize the deviation and accuracy of the sensor. By analyzing historical data and environmental changes, combined with known standard values ​​for comparison and compensation correction, the calibration parameters of the sensor are adjusted and the environment is adapted. The least square method formula is: in is the estimated calibration parameter, N is the design matrix, and m is the target data;

[0031] The self-diagnosis module is used to monitor the real-time operating status of each sensor after automatic calibration in real time, and automatically diagnose the fault type through abnormal detection. Once a fault is detected, the system will generate a repair plan, software adjustment or hardware maintenance suggestion. The repair plan will be pushed to the terminal of the maintenance personnel to guide the operator to quickly handle it. The abnormal detection formula is: Where K(BD) is the posterior probability of a fault occurring, K(DB) is the likelihood of the observed evidence, K(B) is the prior probability of a fault, and K(D) is the total probability of the evidence.

[0032] Furthermore, the workflow of the intelligent collaboration unit includes:

[0033] The redundancy determination module is used to perform correlation analysis on the characteristic data of multiple sensors, calculate the correlation coefficient, thereby understanding the real environmental change response data, and judging whether the characteristic data is redundant. When the data of multiple sensors are abnormal, the redundancy determination is triggered to issue an alarm. The correlation analysis formula is:

[0034] where f abis the Pearson correlation coefficient of the data of sensors a and b′, and are the means of a and b′ respectively;

[0035] The intelligent dispatching system is used to monitor the environmental change response data and energy consumption of the station in real time, predict future environmental changes, dynamically adjust the operating status of power generation equipment, and use intelligent algorithms to continuously optimize the dispatching strategy. The formula for power generation equipment is: Where Z is the total cost, z j is the operating cost of the jth device, a j For the operating status of the equipment, optimize the scheduling strategy formula: Where y(a) is the objective function, y j (a) is the performance index of each device, o j is the weight of the device.

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

[0037] S1. Use multiple sensors to integrate and dynamically adjust data standardization parameters to transmit only feature data.

[0038] S2. Analyze the characteristic data to obtain abnormal sample data and dynamically adjust the abnormality determination threshold;

[0039] S3, performing regular self-calibration according to the abnormality determination threshold, automatically diagnosing and generating a repair plan;

[0040] S4. Perform collaborative anomaly determination based on the characteristic data and intelligently optimize the equipment.

[0041] Furthermore, in step S1, a method of fusing multiple sensors, dynamically adjusting data standardization parameters, and transmitting only feature data is as follows:

[0042] Integrate multiple sensors and use machine learning algorithms to dynamically adjust data standardization parameters based on sensor characteristics and environmental changes to eliminate sensor accuracy differences and obtain standard data.

[0043] Used to pre-process the standard data, transmit only data with key features, and add different communication modes to switch between different environments;

[0044] In step S2, the characteristic data is analyzed to obtain abnormal sample data, and the method for dynamically adjusting the abnormality determination threshold is:

[0045] Used to access and obtain the characteristic data of each new energy station in real time, integrate and process the characteristic data of each station, and realize seamless connection of data from different stations through standardized data formats and protocols;

[0046] The acquired feature data is analyzed, deeply mined and pattern recognized to identify the common features and individual features of the feature data, and further analyzed and trained for abnormality determination thresholds.

[0047] Further, in step S3, the method of periodically performing self-calibration according to the abnormality determination threshold, automatically diagnosing and generating a repair plan is as follows:

[0048] Regularly and automatically calibrate the sensor based on the real-time change of the abnormality determination threshold, and adapt the sensor to the environment;

[0049] Based on the real-time operating status of the sensor after the automatic calibration, automatically diagnose and report the cause of the abnormal state, generate a repair plan, and perform software maintenance or hardware maintenance;

[0050] In step S4, the method of performing collaborative abnormality determination and intelligently optimizing the equipment according to the characteristic data is as follows:

[0051] Through data association analysis and fusion judgment, the characteristic data of multiple sensors are subjected to correlation analysis, so as to understand the real environmental change reaction data, and trigger an alarm when the characteristic data of multiple sensors are abnormal;

[0052] Intelligently dispatch equipment according to the environmental change response data.

[0053] Beneficial Effects

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

[0055] Beneficial effects:

[0056] When used, the present invention is conducive to ensuring data accuracy, reducing bandwidth consumption, improving real-time performance, adapting to different working conditions, reducing the impact of duplicate data, dynamically adjusting the abnormality judgment threshold according to real-time data fluctuations and seasonal changes, avoiding the limitations of traditional static thresholds, ensuring that the accuracy of the sensor is always in the best state, reducing errors caused by environmental changes, improving the reliability and maintenance efficiency of the system, identifying sensor failures, and ensuring through correlation analysis that even if individual sensors fail, the system can still provide reliable data support, which is conducive to reducing energy consumption and improving power generation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1This is a system diagram of a new energy station environment monitoring system based on chemical sensors of the present invention;

[0058] Figure 2 The present invention is a flow chart of a new energy station environment monitoring method based on chemical 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 environment monitoring system based on chemical sensors, including: an edge computing unit, which fuses multiple sensors for use, dynamically adjusts data standardization parameters, and only transmits characteristic data;

[0064] The edge computing unit includes:

[0065] Dynamic integration system integrates multiple sensors and uses machine learning algorithms to dynamically adjust data standardization parameters based on sensor characteristics and environmental changes, eliminating sensor accuracy differences and obtaining standard data.

[0066] An edge mixing module is used to pre-process the standard data, transmit only data with key features, and add different communication modes to switch between different environments;

[0067] The workflow of the edge computing unit includes:

[0068] The dynamic integration system is used to process a large amount of data from multiple sensors, integrating the data of temperature, humidity, air pressure and chemical concentration into a data pool. During the fusion process, the weight of each sensor is updated through the weighted fusion mechanism and the accuracy, stability and environmental changes of each sensor. The weighted average algorithm formula is: where q fin is the fused output, e j is the measurement value of the jth sensor, w j is the weight of the sensor, u is the number of sensors, and the weight update formula is: in is the noise variance of sensor j at time s;

[0069] The edge mixing module is used to analyze and process the standard data, use principal component analysis to reduce the dimension of the standard data, extract the most representative feature data, and perform outlier detection to determine whether the feature data exceeds a reasonable range, screen out the feature data with high correlation, reduce unnecessary data transmission, and use dynamic optimization formulas to reduce bandwidth requirements and improve system real-time performance. Under different environmental conditions, the communication mode switching formula will select the most appropriate communication protocol based on the current bandwidth requirements, delay requirements, and energy consumption constraints. The principal component analysis formula is: W pca =R S (W-β), where W pca is the data after dimensionality reduction, R is the principal component matrix, β is the mean of the data, and the anomaly detection formula is: Where T j is the T value of the jth data, is the observed value, β and α are the mean and standard deviation of the data set respectively, and the dynamic optimization formula is: U eff =U orig × cpn , where U eff is the effective transmission rate, U orig is the original data transmission rate, χ cpn To achieve compression efficiency, the communication mode switching formula is: Minimize y(B, L, P) = m1×B+m2×L+m3×P, where m1, m2, m3 are weight factors used to adjust the relative importance of bandwidth, delay and power consumption, and B, L, P are network bandwidth, delay and power consumption;

[0070] Specifically, a dynamic integration system is used to process data from multiple sensors such as chemical sensors, temperature and humidity sensors, and air pressure sensors, and dynamically adjust the weight of each sensor according to the sensor's accuracy, stability, and environmental changes. At the same time, the edge mixing module performs dimensionality reduction processing on the data from multiple sensors, screens out important features and performs outlier detection to avoid transmitting useless data. When the environment changes, it automatically switches between different communication protocols such as 4G, 5G, Wi-Fi, and NB-IOT to ensure efficient data transmission, which is conducive to ensuring data accuracy, reducing bandwidth consumption, improving real-time performance, and adapting to different working conditions.

[0071] The cross-region access unit performs analysis based on the characteristic data to obtain abnormal sample data and dynamically adjust the abnormal determination threshold;

[0072] Cross-region access units include:

[0073] A distributed data platform is used to access and obtain the characteristic data of each new energy station in real time, integrate and process the characteristic data of each station, and realize seamless connection of data from different stations through standardized data formats and protocols;

[0074] The feature analysis and training module analyzes, deeply mines and recognizes patterns of the acquired feature data, identifies the common features and individual features of the feature data, and further analyzes and trains the abnormality determination threshold;

[0075] The workflow for cross-region access units includes:

[0076] The distributed data platform collects the characteristic data by accessing the database of each station in real time. The characteristic data format and protocol are deduplicated and cleaned through the hash algorithm formula, so that the characteristic data of different stations can ensure data uniqueness. At the same time, in the process of transmitting the characteristic data, the characteristic data of each area is aggregated using the weighted average formula to ensure the accuracy and completeness of the characteristic data. The hash algorithm formula is: I(o)=hash(o), where I(o) is the result of hashing data o. The weighted average formula is: Where P agg is the aggregated data, d j is the data from the jth station, a j is the corresponding weight;

[0077] The feature analysis training module uses big data analysis to conduct in-depth mining of the massive feature data collected across regions, analyze the feature data of each station, identify and extract the common features and individual features of the feature data, and discover the potential laws of environmental changes and sensor failures between stations, so as to dynamically adjust the abnormal judgment threshold. The formula for deep data mining is: Where B is the number of clusters, D b is the number of samples in the b-th cluster, δb is the center of the b-th cluster, and the dynamic anomaly threshold adjustment formula is: ε asd =ε base ×(1+φ), where ε asd is the adjusted threshold, ε base is the initial threshold, φ is the dynamic adjustment factor calculated according to data fluctuations;

[0078] Specifically, data from each new energy station is accessed and collected in real time through a distributed data platform to ensure data uniqueness and avoid duplicate data. Data from each region is aggregated to ensure data accuracy and completeness, which helps reduce the impact of duplicate data. The feature analysis training module is used to conduct in-depth mining of data from multiple stations, extract common and individual features, establish a dynamic anomaly determination model, identify environmental patterns and potential problems in different regions, and dynamically adjust the anomaly determination threshold according to real-time data fluctuations and seasonal changes, which helps avoid the limitations of traditional static thresholds.

[0079] A self-monitoring unit that performs regular self-calibration according to the abnormality determination threshold, automatically diagnoses and generates a repair plan;

[0080] The self-monitoring unit includes:

[0081] An automatic calibration module, which automatically calibrates the sensor regularly based on the real-time change of the abnormality determination threshold value and adapts the sensor to the environment;

[0082] A self-diagnosis module, which automatically diagnoses and reports the cause of the abnormal state based on the real-time operating state of the sensor after the automatic calibration, and generates a repair plan to perform software maintenance or hardware maintenance;

[0083] The workflow of the self-monitoring unit includes:

[0084] The automatic calibration module performs automatic calibration according to the preset cycle, real-time feedback, and real-time changes in the abnormal judgment threshold. In the process of automatic calibration, the least square method is used to optimize the deviation and accuracy of the sensor. By analyzing historical data and environmental changes, combined with known standard values ​​for comparison and compensation correction, the calibration parameters of the sensor are adjusted and the environment is adapted. The least square method formula is: in is the estimated calibration parameter, N is the design matrix, and m is the target data;

[0085] The self-diagnosis module is used to monitor the real-time operating status of each sensor after automatic calibration in real time, and automatically diagnose the fault type through abnormal detection. Once a fault is detected, the system will generate a repair plan, software adjustment or hardware maintenance suggestion. The repair plan will be pushed to the terminal of the maintenance personnel to guide the operator to quickly handle it. The abnormal detection formula is: Where K(BD) is the posterior probability of a fault, K(DB) is the likelihood of the observed evidence, K(B) is the prior probability of a fault, and K(D) is the total probability of the evidence;

[0086] Specifically, the automatic calibration module automatically starts the sensor calibration process according to the preset cycle or real-time feedback, optimizes the sensor's deviation and accuracy, and adjusts the sensor's calibration parameters by comparing historical data and environmental changes to ensure that they are consistent with the standard values. This helps to ensure that the sensor's accuracy is always in the best state and reduce errors caused by environmental changes. The real-time operating data is monitored through the self-diagnosis module. Once an anomaly is detected, the posterior probability is automatically calculated, the possibility of failure is analyzed, and a specific repair plan is generated, which helps to improve the system's reliability and maintenance efficiency.

[0087] An intelligent collaboration unit determines collaboration anomalies based on the characteristic data and intelligently optimizes equipment;

[0088] The intelligent collaborative unit includes:

[0089] The redundancy judgment module performs correlation analysis on the characteristic data of multiple sensors through data association analysis and fusion judgment, so as to understand the real environmental change reaction data, and trigger an alarm when the characteristic data of multiple sensors are abnormal;

[0090] An intelligent scheduling system, which intelligently schedules equipment according to the environmental change response data;

[0091] The workflow of the intelligent collaborative unit includes:

[0092] The redundancy determination module is used to perform correlation analysis on the characteristic data of multiple sensors, calculate the correlation coefficient, thereby understanding the real environmental change response data, and judging whether the characteristic data is redundant. When the data of multiple sensors are abnormal, the redundancy determination is triggered to issue an alarm. The correlation analysis formula is:

[0093] where f ab is the Pearson correlation coefficient of the data of sensors a and b′, and are the means of a and b′ respectively;

[0094] The intelligent dispatching system is used to monitor the environmental change response data and energy consumption of the station in real time, predict future environmental changes, dynamically adjust the operating status of power generation equipment, and use intelligent algorithms to continuously optimize the dispatching strategy. The formula for power generation equipment is: Where Z is the total cost, z j is the operating cost of the jth device, a j For the operating status of the equipment, optimize the scheduling strategy formula: Where y(a) is the objective function, y j (a) is the performance index of each device, o j is the weight of the device;

[0095] Specifically, when multiple sensors perform real-time monitoring in the same environment, the redundant judgment module analyzes whether there is any correlation between the sensor data. When both the temperature and humidity sensors are abnormal, but the pressure sensor readings are normal, the redundant judgment module will trigger an alarm to indicate that the temperature and humidity sensors are faulty, which is helpful for identifying sensor failures. Through correlation analysis, it is ensured that even if individual sensors fail, the system can still provide reliable data support. The intelligent scheduling system obtains environmental data and energy consumption data of new energy stations in real time, predicts weather changes in the next few hours, and dynamically adjusts the operating mode of power generation equipment based on the prediction results. When it is predicted that the temperature will rise, the system can reduce the load of some energy equipment, which is helpful for reducing energy consumption and improving power generation efficiency.

[0096] Embodiment 2:

[0097] like Figure 2 As shown, Example 2 provides a new energy station environment monitoring method based on chemical sensors, which includes the following steps:

[0098] S1. Use multiple sensors to integrate and dynamically adjust data standardization parameters to transmit only feature data.

[0099] S2. Analyze the characteristic data to obtain abnormal sample data and dynamically adjust the abnormality determination threshold;

[0100] S3, performing regular self-calibration according to the abnormality determination threshold, automatically diagnosing and generating a repair plan;

[0101] S4. Perform collaborative anomaly determination based on the characteristic data and intelligently optimize the equipment.

[0102] Furthermore, in step S1, a method of fusing multiple sensors, dynamically adjusting data standardization parameters, and transmitting only feature data is as follows:

[0103] Combine multiple sensors and use machine learning algorithms to dynamically adjust data standardization parameters based on sensor characteristics and environmental changes, eliminating sensor accuracy differences and obtaining standard data.

[0104] Used to pre-process the standard data, transmit only data with key features, and add different communication modes to switch between different environments;

[0105] In step S2, the characteristic data is analyzed to obtain abnormal sample data, and the method for dynamically adjusting the abnormality determination threshold is:

[0106] Used to access and obtain the characteristic data of each new energy station in real time, integrate and process the characteristic data of each station, and realize seamless connection of data from different stations through standardized data formats and protocols;

[0107] The acquired feature data is analyzed, deeply mined and pattern recognized to identify the common features and individual features of the feature data, and further analyzed and trained for abnormality determination thresholds.

[0108] Further, in step S3, the method of periodically performing self-calibration according to the abnormality determination threshold, automatically diagnosing and generating a repair plan is as follows:

[0109] Regularly and automatically calibrate the sensor based on the real-time change of the abnormality determination threshold, and adapt the sensor to the environment;

[0110] Based on the real-time operating status of the sensor after the automatic calibration, automatically diagnose and report the cause of the abnormal state, generate a repair plan, and perform software maintenance or hardware maintenance;

[0111] In step S4, the method of performing collaborative abnormality determination and intelligently optimizing the equipment according to the characteristic data is as follows:

[0112] Through data association analysis and fusion judgment, the characteristic data of multiple sensors are subjected to correlation analysis, so as to understand the real environmental change reaction data, and trigger an alarm when the characteristic data of multiple sensors are abnormal;

[0113] Intelligently dispatch equipment according to the environmental change response data.

[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 environmental monitoring system based on chemical sensors, characterized in that: include: The edge computing unit uses multiple sensors for fusion, dynamically adjusts data standardization parameters, and transmits only feature data; The cross-region access unit performs analysis based on the characteristic data to obtain abnormal sample data and dynamically adjust the abnormal determination threshold; A self-monitoring unit that performs regular self-calibration according to the abnormality determination threshold, automatically diagnoses and generates a repair plan; The intelligent collaboration unit makes collaborative anomaly determinations based on the characteristic data and intelligently optimizes the equipment.

2. According to claim 1, a new energy station environment monitoring system based on chemical sensors is characterized in that: The edge computing unit includes: Dynamic integration system integrates multiple sensors and uses machine learning algorithms to dynamically adjust data standardization parameters based on sensor characteristics and environmental changes, eliminating sensor accuracy differences and obtaining standard data. An edge mixing module is used to pre-process the standard data, transmit only data with key features, and add different communication modes to switch between different environments; Cross-region access units include: A distributed data platform is used to access and obtain the characteristic data of each new energy station in real time, integrate and process the characteristic data of each station, and realize seamless connection of data from different stations through standardized data formats and protocols; The feature analysis training module analyzes, deeply mines and recognizes patterns of the acquired feature data, identifies the common features and individual features of the feature data, and further analyzes and trains the abnormality determination threshold.

3. According to claim 2, a new energy station environment monitoring system based on chemical sensors is characterized in that: The self-monitoring unit includes: An automatic calibration module, which automatically calibrates the sensor regularly based on the real-time change of the abnormality determination threshold value and adapts the sensor to the environment; A self-diagnosis module, which automatically diagnoses and reports the cause of the abnormal state based on the real-time operating state of the sensor after the automatic calibration, and generates a repair plan to perform software maintenance or hardware maintenance; The intelligent collaborative unit includes: The redundancy judgment module performs correlation analysis on the characteristic data of multiple sensors through data association analysis and fusion judgment, so as to understand the real environmental change reaction data, and trigger an alarm when the characteristic data of multiple sensors are abnormal; An intelligent scheduling system intelligently schedules equipment according to the environmental change response data.

4. According to claim 3, a new energy station environment monitoring system based on chemical sensors is characterized in that: The workflow of the edge computing unit includes: The dynamic integration system is used to process data from multiple sensors, integrating the data of temperature, humidity, air pressure and chemical concentration into a data pool. During the fusion process, the weight of each sensor is updated through the weighted fusion mechanism and the accuracy, stability and environmental changes of each sensor. The weighted average algorithm formula is: where q fin is the fused output, e j is the measurement value of the jth sensor, w j is the weight of the sensor, u is the number of sensors, and the weight update formula is: in is the noise variance of sensor j at time s; The edge mixing module is used to analyze and process the standard data, use principal component analysis to reduce the dimension of the standard data, extract the most representative feature data, and perform outlier detection to determine whether the feature data exceeds a reasonable range, screen out the feature data with high correlation, reduce unnecessary data transmission, and use dynamic optimization formulas to reduce bandwidth requirements and improve system real-time performance. Under different environmental conditions, the communication mode switching formula will select the most appropriate communication protocol based on the current bandwidth requirements, delay requirements, and energy consumption constraints. The principal component analysis formula is: W pca =R S (W-β), where W pca is the data after dimensionality reduction, R is the principal component matrix, β is the mean of the data, and the anomaly detection formula is: Where T j is the T value of the jth data, is the observed value, β and α are the mean and standard deviation of the data set respectively, and the dynamic optimization formula is: U eff =U orig × cpn , where U eff is the effective transmission rate, U orig is the original data transmission rate, χ cpn For compression efficiency, the communication mode switching formula is: Minimize y(B, L, P) = m1×B+m2×L+m3×P, where m1, m2, m3 are weight factors used to adjust the relative importance of bandwidth, delay and power consumption, and B, L, P are network bandwidth, delay and power consumption.

5. The new energy station environment monitoring system based on chemical sensors according to claim 4 is characterized in that: The workflow for cross-region access units includes: The distributed data platform collects the characteristic data by accessing the database of each station in real time. The characteristic data format and protocol are deduplicated and cleaned through the hash algorithm formula, so that the characteristic data of different stations can ensure data uniqueness. At the same time, in the process of transmitting the characteristic data, the characteristic data of each area is aggregated using the weighted average formula to ensure the accuracy and completeness of the characteristic data. The hash algorithm formula is: I(o)=hash(o), where I(o) is the result of hashing data o. The weighted average formula is: Where P agg is the aggregated data, d j is the data from the jth station, a j is the corresponding weight; The feature analysis training module uses big data analysis to conduct in-depth mining of the massive feature data collected across regions, analyze the feature data of each station, identify and extract the common features and individual features of the feature data, and discover the potential laws of environmental changes and sensor failures between stations, so as to dynamically adjust the abnormal judgment threshold. The formula for deep data mining is: Where B is the number of clusters, D b is the number of samples in the b-th cluster, δb is the center of the b-th cluster, and the dynamic anomaly threshold adjustment formula is: ε asd =ε base ×(1+φ), where ε asd is the adjusted threshold, ε base is the initial threshold, and φ is the dynamic adjustment factor calculated according to data fluctuations.

6. A new energy station environment monitoring system based on chemical sensors according to claim 5, characterized in that: The workflow of the self-monitoring unit includes: The automatic calibration module performs automatic calibration according to the preset cycle, real-time feedback, and real-time changes in the abnormal judgment threshold. In the process of automatic calibration, the least square method is used to optimize the deviation and accuracy of the sensor. By analyzing historical data and environmental changes, combined with known standard values ​​for comparison and compensation correction, the calibration parameters of the sensor are adjusted and the environment is adapted. The least square method formula is: in is the estimated calibration parameter, N is the design matrix, and m is the target data; The self-diagnosis module is used to monitor the real-time operating status of each sensor after automatic calibration in real time, and automatically diagnose the fault type through abnormal detection. Once a fault is detected, the system will generate a repair plan, software adjustment or hardware maintenance suggestion. The repair plan will be pushed to the terminal of the maintenance personnel to guide the operator to quickly handle it. The abnormal detection formula is: Where K(BD) is the posterior probability of a fault occurring, K(DB) is the likelihood of the observed evidence, K(B) is the prior probability of a fault, and K(D) is the total probability of the evidence.

7. A new energy station environment monitoring system based on chemical sensors according to claim 6, characterized in that: The workflow of the intelligent collaborative unit includes: The redundancy determination module is used to perform correlation analysis on the characteristic data of multiple sensors, calculate the correlation coefficient, thereby understanding the real environmental change response data, and judging whether the characteristic data is redundant. When the data of multiple sensors are abnormal, the redundancy determination is triggered to issue an alarm. The correlation analysis formula is: where f ab is the Pearson correlation coefficient of the data of sensors a and b′, and are the means of a and b′ respectively; The intelligent dispatching system is used to monitor the environmental change response data and energy consumption of the station in real time, predict future environmental changes, dynamically adjust the operating status of power generation equipment, and use intelligent algorithms to continuously optimize the dispatching strategy. The formula for power generation equipment is: Where Z is the total cost, z j is the operating cost of the jth device, a j For the operating status of the equipment, optimize the scheduling strategy formula: Where y(a) is the objective function, y j (a) is the performance index of each device, o j is the weight of the device.

8. A method for monitoring the environment of a new energy station based on a chemical sensor, according to a system for monitoring the environment of a new energy station based on a chemical sensor according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1. Use multiple sensors to integrate and dynamically adjust data standardization parameters to transmit only feature data. S2. Analyze the characteristic data to obtain abnormal sample data and dynamically adjust the abnormality determination threshold; S3, performing regular self-calibration according to the abnormality determination threshold, automatically diagnosing and generating a repair plan; S4. Perform collaborative anomaly determination based on the characteristic data and intelligently optimize the equipment.

9. A new energy station environment monitoring method based on chemical sensors according to claim 8, characterized in that: In step S1, a method of fusing multiple sensors, dynamically adjusting data standardization parameters, and transmitting only feature data is as follows: Integrate multiple sensors and use machine learning algorithms to dynamically adjust data standardization parameters based on sensor characteristics and environmental changes to eliminate sensor accuracy differences and obtain standard data. Used to pre-process the standard data, transmit only data with key features, and add different communication modes to switch between different environments; In step S2, the characteristic data is analyzed to obtain abnormal sample data, and the method for dynamically adjusting the abnormality determination threshold is: Used to access and obtain the characteristic data of each new energy station in real time, integrate and process the characteristic data of each station, and realize seamless connection of data from different stations through standardized data formats and protocols; The acquired feature data is analyzed, deeply mined and pattern recognized to identify the common features and individual features of the feature data, and further analyzed and trained for abnormality determination thresholds.

10. A new energy station environment monitoring method based on chemical sensors according to claim 9, characterized in that: In step S3, the method of periodically performing self-calibration according to the abnormality determination threshold, automatically diagnosing and generating a repair plan is as follows: Regularly and automatically calibrate the sensor based on the real-time change of the abnormality determination threshold, and adapt the sensor to the environment; Based on the real-time operating status of the sensor after the automatic calibration, automatically diagnose and report the cause of the abnormal state, generate a repair plan, and perform software maintenance or hardware maintenance; In step S4, the method of performing collaborative abnormality determination and intelligently optimizing the equipment according to the characteristic data is as follows: Through data association analysis and fusion judgment, the characteristic data of multiple sensors are subjected to correlation analysis, so as to understand the real environmental change reaction data, and trigger an alarm when the characteristic data of multiple sensors are abnormal; Intelligently dispatch equipment according to the environmental change response data.

Images