New energy station operation optimization system and method based on intelligent sensor

Abstract

The invention relates to the technical field of intelligent regulation and control, in particular to a new energy station operation optimization system and method based on an intelligent sensor, and the system comprises a data collection module which is used for collecting the operation data of a wind driven generator and environment data; the data pre-processing module is used for performing data pre-processing on the operation data and the environment data of the wind driven generator according to a data pre-processing method; the analysis and operation module is used for calculating a generator operation state coefficient according to the preprocessed wind power generation operation data; the data transmission module is used for transmitting the analysis operation data; the function prediction module is used for performing power prediction on the analysis operation data to obtain predicted power and performing optimization adjustment on the predicted power according to a standard root-mean-square error value; and the decision execution module is used for performing decision simulation drilling on the state detection signal and performing decision execution. The operation efficiency of the wind driven generator is improved, and stable operation of the generator is ensured.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to a new energy station operation optimization system and method based on intelligent sensors. Background Art

[0002] With the continuous development of intelligent technology, wind storage power stations, as an important part of renewable energy power generation, have great significance for the safe and stable operation of the power system in terms of operational stability and reliability. However, the maximum power operation mode of the wind farm will weaken the frequency stability and recovery ability of the power grid under the impact of large power shortages. This method can monitor the frequency and voltage changes of the power grid in real time, and adjust the output power of the wind storage power station according to the changes, so as to achieve effective regulation of the frequency and voltage of the power grid. In addition, this method can also make full use of the advantages of the energy storage system to achieve rapid storage and release of energy, and further improve the frequency and voltage regulation capabilities of the wind storage power station. In addition, the system can also realize remote management and control of new energy stations through remote monitoring and intelligent scheduling, and further improve the operating efficiency and economic benefits of the stations.

[0003] Chinese Patent Publication No.: CN116073434A discloses a frequency regulation and voltage regulation control system and method for a wind power station. The present invention includes: a station-level EMS energy management system, a wind turbine energy management system, an electric hydrogen production power management system, and an inertia management system; the whole station unit architecture is controlled according to the station-level EMS energy management system, and the inertia response control, primary frequency regulation control, and voltage regulation control of the wind power station are realized through the domain controller. The maximum power operation mode of the wind farm weakens the frequency stability and recovery ability of the power grid under the impact of large power shortages. The response adjustment is differentiated according to the priority, and the EMS system is established through the programmable domain controller to realize unified management and control. The structure includes a supercapacitor group, an electric hydrogen production device, a wind turbine and its inverter, a transformer and a domain controller, which are connected in series and connected to a 35kV bus. This solution lacks fault detection and maintenance functions, and cannot promptly detect and handle abnormal conditions in the operation of the wind power station, resulting in low operating efficiency of wind turbines. Summary of the invention

[0004] To this end, the present invention provides a new energy station operation optimization system and method based on intelligent sensors to overcome the problem in the prior art that due to the lack of fault detection and maintenance functions, abnormal conditions in the operation of wind power stations cannot be discovered and processed in time, resulting in low operating efficiency of wind turbines.

[0005] To achieve the above objectives, on the one hand, the present invention provides a new energy station operation optimization system based on intelligent sensors, the system comprising: A data collection module is used to collect wind turbine operation data and environmental data; A data preprocessing module, used to perform data preprocessing on the wind turbine operation data and environmental data according to a data preprocessing method to obtain preprocessed wind turbine operation data and preprocessed environmental data; An analysis and calculation module, used to calculate the generator operation state coefficient according to the wind speed data in the preprocessed wind power generation operation data, and to perform fault detection and repair on the generator according to the generator operation state coefficient, and also used to calculate the environmental state parameters according to the preprocessed environmental data, and to judge the state of the preprocessed environmental data according to the environmental state parameters, and to optimize the generator operation state coefficient when the state of the preprocessed environmental data is abnormal, and also used to judge the effectiveness of manual intervention according to the ambient temperature in the preprocessed environmental data, and to optimize the environmental state parameters according to the result of the judgment of the effectiveness of manual intervention; A data transmission module, used to use the pre-processed wind power generation operation data, the generator operation state and the optimized environmental state parameters as analysis operation data, and transmit the analysis operation data; A power prediction module is used to perform power prediction on the analysis and operation data to obtain the predicted power, and is also used to detect and repair the generator according to the predicted power, and is also used to calculate the standard root mean square error value and optimize and adjust the predicted power according to the standard root mean square error value; The decision execution module is used to integrate and analyze the analytical operation data according to the intelligent sensor to obtain the status detection signal, and is also used to perform decision simulation exercises on the status detection signal, and to execute the decision on the predicted power according to the decision simulation exercise results.

[0006] Furthermore, when analyzing the operating status of the generator, the analysis and calculation module calculates the generator operating status coefficient S according to the wind speed data V in the preprocessed wind power generation operating data, the implementation power P corresponding to the wind speed in the selected time period, the wind speed data point i in the time period and the standard power Pr corresponding to the wind speed data point, and sets , where n is the threshold of wind speed data point i.

[0007] Furthermore, when the analysis and calculation module determines the type of the generator operating state, the generator operating state coefficient S is compared with a preset generator operating state coefficient standard value S0, the type of the generator operating state is determined according to the comparison result, and the generator is fault detected and repaired according to the determination result, wherein: When S≥S0, the generator operation status is determined to be normal, and no fault detection or maintenance is performed on the generator; When S<S0, the type of the generator operation state is determined to be abnormal, and the generator is fault detected and repaired.

[0008] Furthermore, when optimizing the operating state of the generator, the analysis and calculation module calculates the environmental state parameter E according to the pre-processed ambient temperature A, ambient humidity B and ambient pressure C, sets E=0.4A+0.3B+0.3C, compares the environmental state parameter E with the preset environmental state parameter standard value E0, judges the state of the environmental data according to the comparison result, and optimizes the generator operating state coefficient S according to the judgment result, wherein: When E≥E0, the analysis and operation module determines that the state of the preprocessed environmental data is normal, and does not optimize the generator operation state coefficient S; When E<E0, the analysis and operation module determines that the state of the pre-processed environmental data is abnormal, optimizes the generator operation state coefficient S, and calculates the optimization coefficient α, setting α=0.6×e -0.02×E +0.4×(0.002×E+0.65） , e is a natural number base, the generator operating state coefficient S is optimized according to the optimization coefficient α, the optimized generator operating state coefficient is Sα, and Sα=α×S is set.

[0009] Furthermore, when optimizing the environmental state parameter E, the analysis and operation module compares the ambient temperature A in the preprocessed environmental data with the preset ambient temperature AO, judges the effectiveness of manual intervention according to the comparison result, and optimizes the environmental state parameter E according to the judgment result of the effectiveness of manual intervention, wherein: When A≤A0, the effectiveness of the manual intervention is determined to be passed, and the analysis and calculation module does not optimize the environmental state parameter E; When A>A0, the effectiveness of the manual intervention is judged as not passed, and the analysis and calculation module optimizes the environmental state parameter E and calculates the environmental impact coefficient θ, setting θ=0.66+e -0.1×[(E0-E)+12]）] , e is a natural number base, the environmental state parameter E is optimized according to the environmental impact coefficient θ, the optimized environmental state parameter is Ee, and Ee=θ×E is set.

[0010] Furthermore, when the power prediction module performs power prediction on the analysis and operation data, the analysis and operation data is input into the power prediction model to construct a power prediction model, wherein the power prediction database is divided into a 75% prediction training set, a 15% prediction verification set and a 10% prediction test set, the prediction training set is input into the decision tree structure model to train the decision tree structure model, and the prediction verification set is input into the trained decision tree structure model, the decision tree structure model is iteratively optimized for hyperparameters, and the prediction test set is input into the iteratively optimized decision tree structure model to perform power test on the iteratively optimized decision tree structure model to obtain predicted power, the total number of predicted power samples is set to g0, the number of correctly predicted test samples is g, the predicted power accuracy is G, G=g / g0, the predicted power accuracy G is compared with the preset predicted power accuracy G0, the training compliance of the iteratively optimized decision tree structure model is judged according to the comparison result, and the judgment result is output, wherein: When G≥G0, it is determined that the iteratively optimized decision tree structure model training has reached the standard. The iteratively optimized decision tree structure model is output as a power prediction model; When G<G0, it is determined that the training of the iteratively optimized decision tree structure model does not meet the standards, the power prediction database is updated to obtain an updated power prediction database, and the decision tree model is trained, hyperparameters are iteratively optimized and power tested according to the updated power prediction database until the training of the decision tree structure model meets the standards.

[0011] Furthermore, when judging the predicted power, the power prediction module calculates the predicted power Py according to the radius r, wind speed v, air density ρ and efficiency factor n of the wind turbine, and sets Py= , and compare the predicted power Py with the preset power Py0, and judge the predicted power Py according to the comparison result, where: When Py≥Py0, the power prediction module determines that the predicted power Py passes the power prediction and can continue to run; When Py<Py0, the power prediction module determines that the predicted power Py does not pass the power prediction and cannot continue to operate, and needs to be tested and repaired again.

[0012] Furthermore, when the power prediction module optimizes and adjusts the predicted power Py, the standard root mean square error value R is calculated according to the wind power operation power and the power prediction value, and R= ,in: is the actual value of wind power operation power, is the power prediction value, N is the actual total number of samples, and i is the actual wind power operation rate; The power prediction module compares the standard root mean square error value R with the preset standard root mean square error value R0, judges the error range of the predicted power Py according to the comparison result, and optimizes and adjusts the predicted power Py according to the judgment result, wherein: When R>R0, the power prediction module determines that the error range of the predicted power Py is unacceptable and does not pass the power prediction, optimizes and adjusts the predicted power Py, and sets the error adjustment coefficient β to optimize and adjust the predicted power Py, setting β=e -[(R-R0)+0.7] +0.15, e is the base of the natural logarithm, and the predicted power after optimization and adjustment is Py2=β×Py; When R≤R0, the power prediction module determines that the error range of the predicted power Py is within an acceptable range and passes the power prediction without optimizing and adjusting the predicted power Py.

[0013] Furthermore, when the decision execution module performs decision execution on the state detection signal, the decision execution module performs decision simulation drill on the state detection signal through the signal simulation drill model, and performs decision execution on the predicted power according to the drill result, and constructs a signal simulation drill model, wherein the simulation drill database is divided into a 70% drill training set, a 20% drill verification set and a 10% drill test set, the drill training set is input into the decision tree model to train the decision tree model, and the drill verification set is input into the trained decision tree model, and the trained The decision tree model is iteratively optimized for hyperparameters, and the drill test set is input into the iteratively optimized decision tree model to perform a drill test on the iteratively optimized decision tree model to obtain the drill test results. The total number of samples in the drill test set is set to Y0, the number of correct drill test samples is set to Y1, and the drill test accuracy is set to Y, Y=Y1 / Y0. The drill test accuracy Y is compared with the preset drill test accuracy Y0. According to the comparison results, the training compliance of the iteratively optimized decision tree model is judged, and the judgment results are output, where: When Y≥Y0, it is determined that the iteratively optimized decision tree model is successfully trained, a decision is executed on the predicted power, and the iteratively optimized decision tree model is output as a signal simulation exercise model; When Y<Y0, it is determined that the training of the iteratively optimized decision tree model has failed, and the simulation exercise database is updated to obtain an updated simulation exercise database. The decision tree model is trained, hyperparameters are iteratively optimized, and exercised and tested according to the updated simulation exercise database until the decision tree model is successfully trained.

[0014] On the other hand, the present invention also provides a new energy station operation optimization method based on intelligent sensors, the method comprising: Step S1, collecting wind turbine operation data and environmental data; Step S2, performing data preprocessing on the wind turbine operating data and the environmental data according to the data preprocessing method to obtain preprocessed wind turbine operating data and preprocessed environmental data; Step S3, calculating the generator operation state coefficient according to the wind speed data in the preprocessed wind power generation operation data, and performing fault detection and maintenance on the generator according to the generator operation state coefficient; Step S4, calculating the environmental state parameters according to the preprocessed environmental data, judging the state of the preprocessed environmental data according to the environmental state parameters, and optimizing the generator operation state coefficient when the state of the preprocessed environmental data is abnormal; Step S5, judging the effectiveness of manual intervention according to the ambient temperature in the preprocessed environmental data, and optimizing the environmental state parameters according to the result of judging the effectiveness of manual intervention; Step S6, using the pre-processed wind power generation operation data, the generator operation state and the optimized environmental state parameters as analysis operation data, and transmitting the analysis operation data; Step S7, performing power prediction on the analyzed operation data to obtain predicted power, which is also used to detect and repair the generator according to the predicted power; Step S8, calculating a standard root mean square error value according to the wind power operation power and the power prediction value, and optimizing and adjusting the predicted power according to the standard root mean square error value; Step S9, integrating and analyzing the analysis operation data according to the intelligent sensor to obtain a state detection signal; Step S10, performing a decision simulation exercise on the state detection signal, and executing a decision on the predicted power according to the decision simulation exercise result.

[0015] Compared with the prior art, the beneficial effect of the present invention is that the system collects the operating data and environmental data of the wind turbine generator through the data collection module, provides data support for subsequent data preprocessing, thereby ensuring the comprehensiveness and completeness of the data preprocessing process, the system removes invalid and abnormal data in the collected data through the data preprocessing module, and unifies the data format and unit, and further analyzes the efficiency and accuracy of the operation process, the system judges the type of the generator operation status through the analysis operation module, and can timely discover and locate the fault of the wind turbine generator, thereby reducing the downtime and maintenance costs caused by the fault, improving the reliability and safety of the generator, and providing accurate and comprehensive analysis operation data for power prediction and decision execution, and the system ensures that the analysis operation data can be transmitted to the power prediction module and the decision execution module in real time and accurately through the data transmission module. block, protecting the integrity and security of data during transmission, further improving the efficiency of data transmission, and reducing communication delays. The system accurately predicts the power of the generator through the power prediction module, providing a reliable basis for power generation scheduling and resource optimization. By comparing with the preset value, potential generator failures can be discovered in time and maintenance can be carried out in advance. The predicted power is optimized and adjusted according to the standard root mean square error value, thereby improving the accuracy and practicality of the prediction. The system makes intelligent decisions based on intelligent sensors and integrated analysis data through the decision execution module, verifies the feasibility and effectiveness of the decision plan through decision simulation exercises, responds to and executes the status detection signal in real time, and makes decisions on the predicted power according to the results of the decision simulation exercises, continuously optimizes the operating status and efficiency of the generator, improves the accuracy and efficiency of the decision, and ensures that the generator can operate stably.

[0016] In particular, in the analysis and calculation module, when analyzing the operating status of the generator, the generator operating status coefficient S is calculated and compared with the preset generator operating status standard value S0 to achieve real-time monitoring and accurate judgment of the generator operating status, timely discover and deal with the generator fault, improve the reliability and safety of the generator, reduce downtime and maintenance costs caused by faults, and at the same time improve power generation efficiency and economic benefits.

[0017] In particular, in the analysis and calculation module, when optimizing the operating status of the generator, the intelligent optimization and refined adjustment of the operating status of the generator are achieved through the calculation and comparison of pre-processed environmental data, environmental status parameters, and calculation and adjustment of optimization coefficients, thereby improving the stability of the system, power generation efficiency, and maintenance cost-effectiveness.

[0018] In particular, in the operation and analysis module, when optimizing the environmental state parameters, the optimization of the environmental state parameters is achieved through the synergistic effect of the analysis operation module, the effectiveness judgment of manual intervention and the environmental impact coefficient θ, thereby improving the accuracy and efficiency of the optimization and enhancing the environmental adaptability of the system.

[0019] In particular, in the data transmission module, when the analysis and operation data are transmitted, the interconnection and transmission of the analysis and operation data are realized by selecting the communication protocol, which improves the efficiency and security of data transmission and provides strong support for power prediction.

[0020] In particular, in the power prediction module, when performing power prediction on the analysis and operation data, power prediction automation is achieved by constructing a power prediction model, database partitioning, model training, hyperparameter optimization and result evaluation, reducing the effectiveness of manual intervention, and improving the model's adaptability to new data and the accuracy of power prediction.

[0021] In particular, in the power prediction module, when judging the predicted power of the wind turbine, through accurate calculation and comparison of the predicted power, real-time monitoring and judgment of the operating status of the wind turbine can be achieved, thereby improving the operating efficiency and safety of the wind turbine.

[0022] In particular, in the power prediction module, when optimizing and adjusting the wind power prediction, the root mean square error is calculated and compared with the preset standard, and the predicted power is optimized and adjusted according to the comparison result, so as to accurately judge the error range of the predicted power and improve the accuracy and efficiency of the wind power prediction.

[0023] In particular, in the decision-making execution module, when the analytical calculation data is integrated and analyzed, the application of intelligent sensors and fault diagnosis models can achieve accurate monitoring of the operating status and fault diagnosis of wind turbines, thereby improving the operating efficiency, operation and maintenance management level and economic benefits of wind turbines.

[0024] In particular, in the decision execution module, when executing decisions on status detection signals, a decision tree model is constructed through a signal simulation exercise model to divide the data set, train the model, iteratively optimize the hyperparameters, and conduct exercise tests. The optimized decision tree model executes decisions on the predicted power, realizes decision simulation exercises on status detection signals, and trains and optimizes the decision tree model, thereby improving the accuracy and efficiency of decision execution as well as the operating efficiency and management level of the entire system. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a structural diagram of the new energy station operation optimization system based on smart sensors in this embodiment; Figure 2 It is a flowchart of the new energy station operation optimization method based on smart sensors in this embodiment. DETAILED DESCRIPTION

[0026] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0027] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0028] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0029] See also Figure 1 As shown, it is a structural diagram of the new energy station operation optimization system based on intelligent sensors in this embodiment, and the system includes: A data collection module is used to collect wind turbine operation data and environmental data; A data preprocessing module, used to perform data preprocessing on the wind turbine operation data and the environmental data according to the data preprocessing method to obtain the preprocessed wind turbine operation data and the preprocessed environmental data, and the data preprocessing module is connected to the data collection module; An analysis and operation module is used to calculate the generator operation state coefficient according to the wind speed data in the preprocessed wind power generation operation data, and to perform fault detection and maintenance on the generator according to the generator operation state coefficient, and is also used to calculate the environmental state parameters according to the preprocessed environmental data, and to judge the state of the preprocessed environmental data according to the environmental state parameters, and to optimize the generator operation state coefficient when the state of the preprocessed environmental data is abnormal, and is also used to judge the effectiveness of manual intervention according to the ambient temperature in the preprocessed environmental data, and to optimize the environmental state parameters according to the judgment result of the effectiveness of manual intervention, and the analysis and operation module is connected to the data preprocessing module; A data transmission module, used to use the pre-processed wind power generation operation data, the generator operation state and the optimized environmental state parameters as analysis and operation data, and transmit the analysis and operation data, the data transmission module is connected to the analysis and operation module; A power prediction module is used to perform power prediction on the analysis and operation data to obtain the predicted power, and is also used to detect and repair the generator according to the predicted power, and is also used to calculate the standard root mean square error value according to the wind power operation power and the power prediction value, and optimize and adjust the predicted power according to the standard root mean square error value. The power operation module is connected with the operation analysis module and the data transmission module; The decision execution module is used to integrate and analyze the analysis and operation data according to the intelligent sensor to obtain the status detection signal, and is also used to perform decision simulation exercises on the status detection signal, and to execute the decision on the predicted power according to the decision simulation exercise results. The decision execution module is connected with the power prediction module, the data transmission module and the operation analysis module.

[0030] Specifically, the system is applied to the intelligent decision-making and execution terminal of new energy wind power. By integrating data collection, preprocessing, analysis and calculation, transmission, power prediction and decision-making execution modules, it realizes the efficient, stable and sustainable operation of the integrated control of new energy generators and improves the operating efficiency of wind generators. The system collects the operating data and environmental data of wind generators through the data collection module, and provides data support for subsequent data preprocessing, thereby ensuring the comprehensiveness and completeness of the data preprocessing process. The system removes invalid and abnormal data from the collected data through the data preprocessing module, and unifies the data format and unit, and further analyzes the efficiency and accuracy of the calculation process. The system judges the type of generator operating status through the analysis and calculation module, and can timely discover and locate wind generator faults, thereby reducing downtime and maintenance costs caused by faults, improving the reliability and safety of generators, and providing accurate and comprehensive analysis and calculation data for power prediction and decision execution. The system ensures that the analysis and operation data can be transmitted to the power prediction module and the decision execution module in real time and accurately through the data transmission module, protects the integrity and security of the data during the transmission process, further improves the efficiency of data transmission, and reduces communication delays. The system accurately predicts the power of the generator through the power prediction module, providing a reliable basis for power generation scheduling and resource optimization. By comparing with the preset value, potential generator failures can be discovered in time and repaired in advance. The predicted power is optimized and adjusted according to the standard root mean square error value, thereby improving the accuracy and practicality of the prediction. The system makes intelligent decisions based on intelligent sensors and integrated analysis data through the decision execution module, verifies the feasibility and effectiveness of the decision plan through decision simulation exercises, responds to and executes the status detection signal in real time, and makes decisions on the predicted power according to the results of the decision simulation exercises. The operating status and efficiency of the generator are continuously optimized, and the accuracy and efficiency of the decision are improved, thereby ensuring that the generator can operate stably.

[0031] Specifically, when collecting the wind turbine operating data and environmental data, the data collection module is connected to the wind turbine main control PLC through the wind speed sensor, and collects the wind turbine operating data in a multi-threaded manner. The wind turbine operating data includes wind speed data and real-time power of the generator set. The environmental data is collected through temperature sensors, humidity sensors and pressure sensors. The environmental data includes ambient temperature, ambient humidity and ambient pressure. The wind turbine operating data and the environmental data are read using a single-chip microcomputer, and the read data is stored in the cloud.

[0032] Specifically, the wind turbine operation data refers to the parameters and information generated during the operation process, the wind turbine main control PLC communication connection refers to the communication connection between the wind turbine main control PLC and other devices, the multi-threading method refers to the wind turbine operation data programming technology, the wind speed data refers to the air flow speed data, the generator real-time power refers to the output power of the wind turbine at the current moment, the temperature sensor refers to a sensor device for measuring ambient temperature, the humidity sensor refers to a sensor device for measuring ambient humidity, the pressure sensor refers to a sensor device for measuring ambient atmospheric pressure, the ambient temperature refers to the temperature of the wind turbine operating environment, the ambient humidity refers to the humidity of the wind turbine operating environment, the ambient pressure refers to the atmospheric pressure of the wind turbine operating environment, the single-chip microcomputer refers to an integrated circuit chip, and the cloud refers to a software platform that uses application virtualization technology.

[0033] Specifically, in the data collection module, when collecting wind turbine operation data and environmental data, centralized management and remote access of data are achieved through the application of wind speed sensors, communication connections of wind turbine main control PLCs, temperature sensors, humidity sensors, air pressure sensors and single-chip microcomputers, thereby improving the efficiency and accuracy of data collection.

[0034] Specifically, when the data preprocessing module performs data preprocessing on the wind turbine operation data and the environmental data, the data preprocessing method is used to preprocess the wind turbine operation data and the environmental data. The data preprocessing method includes missing value processing, outlier processing, data standardization processing and data normalization processing, and the StandardScaler and MinMaxScaler tools in the sklearn.preprocessing library of Python are used to perform data standardization and normalization processing to obtain the preprocessed wind power generation operation data and the preprocessed environmental data.

[0035] Specifically, the data preprocessing refers to the process of processing the original data before it is further analyzed and modeled, the missing value processing refers to the process of filling and deleting missing and empty values ​​in the original data, the outlier processing refers to the values ​​in the data that obviously deviate from the normal range, the data standardization refers to scaling the data according to the rules so that the data conforms to the distribution characteristics, the normalization processing refers to scaling the data according to the proportion so that the data falls within a specific range, the Python refers to a high-level programming language, the sklearn.preprocessing library refers to a module for preprocessing data, the StandardScaler tool refers to a tool for standardizing data to unit variance and zero mean, the MinMaxScaler refers to a tool for normalizing data, the preprocessed wind power generation operation data refers to the wind power generation operation data after missing value processing, outlier processing, standardization processing and normalization processing, and the preprocessed environmental data refers to the data obtained after preprocessing the environmental monitoring data.

[0036] Specifically, in the data preprocessing module, when preprocessing the wind turbine operation data and environmental data, effective preprocessing of the wind turbine operation data and environmental data is achieved through data preprocessing methods and tools in Python's sklearn.preprocessing library, thereby improving the quality and consistency of the data and thus improving the performance and accuracy of the model.

[0037] Specifically, when analyzing the operating status of the generator, the analysis and calculation module calculates the generator operating status coefficient S according to the wind speed data V in the preprocessed wind power generation operation data, the implementation power P corresponding to the wind speed in the selected time period, the wind speed data point i in the time period and the standard power Pr corresponding to the wind speed data point, and sets , where n is the threshold of wind speed data point i; When the analysis and calculation module determines the type of the generator operating state, the generator operating state coefficient S is compared with a preset generator operating state coefficient standard value S0, the type of the generator operating state is determined according to the comparison result, and the generator is fault detected and repaired according to the determination result, wherein: When S≥S0, the generator operation status is determined to be normal, and no fault detection or maintenance is performed on the generator; When S<S0, the type of the generator operation state is determined to be abnormal, and the generator is fault detected and repaired.

[0038] Specifically, the generator operating status refers to the parameters and performance of the generator during operation, the wind speed data V refers to the measured value of the wind speed in the generator environment, the implemented power P refers to the actual output power of the generator calculated by the wind speed data, the wind speed data point i refers to the specific measurement value and recording point of the wind speed data, the standard power Pr corresponding to the wind speed data point refers to the theoretical standard power output calculated by the generator design specifications and historical data, the preset generator operating status coefficient standard value S0 refers to the standard value used to judge whether the generator operating status is normal, the generator refers to a mechanical equipment that converts mechanical energy into electrical energy, and the inspection and maintenance refer to measures to ensure the normal operation and extend the service life of the generator.

[0039] Specifically, in the analysis and calculation module, when analyzing the operating status of the generator, the generator operating status coefficient S is calculated and compared with the preset generator operating status standard value S0 to achieve real-time monitoring and accurate judgment of the generator operating status, timely discover and deal with the generator fault, improve the reliability and safety of the generator, reduce downtime and maintenance costs caused by faults, and improve power generation efficiency and economic benefits.

[0040] Specifically, when optimizing the operating state of the generator, the analysis and calculation module calculates the environmental state parameter E according to the pre-processed ambient temperature A, ambient humidity B and ambient pressure C, sets E=0.4A+0.3B+0.3C, compares the environmental state parameter E with the preset environmental state parameter standard value E0, judges the state of the environmental data according to the comparison result, and optimizes the generator operating state coefficient S according to the judgment result, wherein: When E≥E0, the analysis and operation module determines that the state of the preprocessed environmental data is normal, and does not optimize the generator operation state coefficient S; When E<E0, the analysis and operation module determines that the state of the pre-processed environmental data is abnormal, optimizes the generator operation state coefficient S, and calculates the optimization coefficient α, setting α=0.6×e -0.02×E +0.4×(0.002×E+0.65） , e is a natural number base, the generator operating state coefficient S is optimized according to the optimization coefficient α, the optimized generator operating state coefficient is Sα, and Sα=α×S is set.

[0041] Specifically, the temperature data A refers to the temperature information in the generator operating environment, which is used to calculate the environmental state parameters after preprocessing. The humidity data B refers to the humidity information in the generator operating environment, which is used to calculate the environmental state parameters after preprocessing. The air pressure data C refers to the air pressure information in the generator operating environment, which is used to calculate the environmental state parameters after preprocessing. The environmental state parameter E refers to a composite parameter used to evaluate the overall state of the generator operating environment. The preset environmental state parameter standard value E0 refers to a threshold for judging whether the environmental state is normally set. This embodiment does not limit the specific value of the preset environmental state parameter standard value E0. Relevant technical personnel in this field can freely set it according to actual conditions, as long as the requirement for comparison with the environmental state parameter E is met. For example, the specific value of the preset environmental state parameter standard value can be set to EO=0.95. The optimization coefficient α refers to a coefficient used to adjust the generator operating state. The optimized generator operating state coefficient Sα refers to the generator operating state adjusted by the optimization coefficient α.

[0042] Specifically, in the analysis and calculation module, when optimizing the operating status of the generator, the intelligent optimization and refined adjustment of the operating status of the generator are achieved through the calculation and comparison of pre-processed environmental data, environmental status parameters, and calculation and adjustment of optimization coefficients, thereby improving the stability of the system, power generation efficiency and maintenance cost-effectiveness.

[0043] Specifically, when optimizing the environmental state parameter E, the analysis and operation module compares the ambient temperature A in the preprocessed environmental data with the preset ambient temperature AO, judges the effectiveness of manual intervention according to the comparison result, and optimizes the environmental state parameter E according to the judgment result, wherein: When A≤A0, the effectiveness of the manual intervention is determined to be effective, and the analysis and calculation module does not optimize the environmental state parameter E; When A>A0, the effectiveness of the human intervention is determined to be invalid, and the analysis and calculation module optimizes the environmental state parameter E and calculates the environmental impact coefficient θ, setting θ=0.66+e -0.1×[(E0-E)+12]）] , e is a natural number base, the environmental state parameter E is optimized according to the environmental impact coefficient θ, the optimized environmental state parameter is Ee, and Ee=θ×E is set.

[0044] Specifically, the preset ambient temperature AO refers to a preset ambient temperature reference value. This embodiment does not limit the standard value of the preset ambient temperature AO. Relevant technical personnel in the field can freely set it according to actual conditions. It only needs to meet the requirements of comparing with the ambient temperature A in the preprocessed environmental data. For example, the standard value of the preset ambient temperature AO can be set to 25°C. The effectiveness of manual intervention refers to the subjective evaluation and decision of the environmental state and analysis results made by human experts and the system based on the preprocessed environmental data. This embodiment does not specifically limit the influencing factors of the effectiveness of manual intervention. Relevant technical personnel in the field can freely set it according to actual conditions. It only needs to meet the requirements of the judgment results of the effectiveness of manual intervention. For example, the influencing factors of the effectiveness of manual intervention can be set to ambient temperature. The environmental impact coefficient θ refers to the coefficient used to measure the impact of the environmental state parameters on the environment after optimization. The optimized environmental state parameter Ee refers to the environmental state parameter optimized by the analysis and operation module.

[0045] Specifically, in the operation and analysis module, when optimizing the environmental state parameters, the optimization of the environmental state parameters is achieved by analyzing the synergistic effect of the operation module, the effectiveness judgment of manual intervention and the environmental impact coefficient θ, thereby improving the accuracy and efficiency of the optimization and enhancing the environmental adaptability of the system.

[0046] Specifically, when the data transmission module transmits the analysis and operation data, it transmits the analysis and operation data through Modbus TCP / IP communication. The analysis and operation data includes pre-processed wind power generation operation data, the generator operation status and the optimized environmental status parameters. The Modbus protocol is used to interconnect the analysis and operation data and transmit them to the power prediction module.

[0047] Specifically, the Modbus TCP / IP communication refers to a variation of the communication protocol for connecting electronic devices based on Ethernet, and the Modbus protocol refers to a communication protocol for connecting generator equipment and realizing data exchange.

[0048] Specifically, in the data transmission module, when the analysis and operation data is transmitted, the communication protocol is selected to realize the interconnection and transmission of the analysis and operation data, thereby improving the efficiency and security of data transmission and providing strong support for power prediction.

[0049] Specifically, when the power prediction module performs power prediction on the analysis and operation data, the analysis and operation data is input into the power prediction model to construct a power prediction model, wherein the power prediction database is divided into a 75% prediction training set, a 15% prediction verification set and a 10% prediction test set, the prediction training set is input into the decision tree structure model to train the decision tree structure model, and the prediction verification set is input into the trained decision tree structure model, the decision tree structure model is iteratively optimized for hyperparameters, and the prediction test set is input into the iteratively optimized decision tree structure model to perform power test on the iteratively optimized decision tree structure model to obtain predicted power, the total number of predicted power samples is set to g0, the number of correctly predicted test samples is g, the predicted power accuracy is G, G=g / g0, the predicted power accuracy G is compared with the preset predicted power accuracy G0, the training compliance of the iteratively optimized decision tree structure model is judged according to the comparison result, and the judgment result is output, wherein: When G≥G0, it is determined that the iteratively optimized decision tree structure model training has reached the standard. The iteratively optimized decision tree structure model is output as a power prediction model; When G<G0, it is determined that the training of the iteratively optimized decision tree structure model does not meet the standards, the power prediction database is updated to obtain an updated power prediction database, and the decision tree model is trained, hyperparameters are iteratively optimized and power tested according to the updated power prediction database until the training of the decision tree structure model meets the standards.

[0050] Specifically, the power prediction model refers to a model for predicting the power generation of renewable energy in the future, the power prediction data set refers to the historical data and real-time data used to store and manage the power prediction of renewable energy power stations, the prediction training set refers to the data set used to train the prediction model, the prediction verification set refers to the data set used to verify the performance of the prediction model, the prediction test set refers to the last set of data that is completely independent of the training set and the verification set, the decision tree structure model refers to the tree structure that describes the classification of instances, the hyperparameter iterative optimization refers to the process of training the model multiple times and adjusting the hyperparameters of the parameters used in the machine learning algorithm, the power test refers to the process of testing and verifying the power output of the generator equipment system, the predicted power refers to an indicator used to evaluate the power output and demand capacity of the generator equipment system in a specific time in the future, and the total number of predicted power samples g0 refers to the total number of samples used to evaluate the performance of the power prediction model, the number of correctly predicted test samples g refers to the number of samples correctly predicted by the model in the test set, the predicted power accuracy G refers to the proportion of power correctly predicted by the model on the test set, the preset predicted power accuracy G0 refers to the predicted power accuracy target value set according to requirements and standards in actual applications and research, the training compliance status refers to whether the model meets the preset performance indicators and goals during the training process, and the update refers to modifying and supplementing the data and functions in the power prediction database. This embodiment does not limit the method for updating the power prediction database. Relevant technical personnel in this field can freely set it according to actual conditions, and only need to meet the needs of the updated power prediction database for training the decision tree structure model, iterative optimization of hyperparameters and power prediction testing. For example, the update method can be set to big data algorithm update.

[0051] Specifically, in the power prediction module, when performing power prediction on the analysis and operation data, power prediction automation is achieved by constructing a power prediction model, database partitioning, model training, hyperparameter optimization and result evaluation, reducing the effectiveness of manual intervention, and improving the model's adaptability to new data and the accuracy of power prediction.

[0052] Specifically, when judging the predicted power, the power prediction module calculates the predicted power Py according to the radius r, wind speed v, air density ρ and efficiency factor n of the wind turbine, and sets Py= , and compare the predicted power Py with the preset power Py0, and judge the predicted power Py according to the comparison result, where: When Py≥Py0, the power prediction module determines that the predicted power Py passes the power prediction and can continue to run; When Py<Py0, the power prediction module determines that the predicted power Py does not pass the power prediction and cannot continue to operate, and needs to be tested and repaired again.

[0053] Specifically, the radius r of the wind turbine refers to the radius of the circle swept by the blades of the wind turbine, the wind speed v refers to the speed at which the wind blows through the blades of the wind turbine, the air density ρ refers to the mass of air per unit volume, the efficiency factor n refers to the actual efficiency of the wind turbine in converting the captured wind energy into electrical energy, the predicted power Py refers to the electric power output by the wind turbine, the predicted power Py includes the wind power operating power and the power prediction value, the wind power operating power refers to the electric power output power generated by the wind turbine and the wind farm during the actual operation, the power prediction value refers to the wind power output power predicted by the power prediction module based on wind speed, wind direction and temperature, the preset pass power Py0 refers to the electric power that the wind turbine can generate under specific conditions, this embodiment does not limit the size range of the preset pass power Py0, relevant technical personnel in this field can freely set it according to actual conditions, and only need to meet the requirements of comparing the predicted power Py with the preset pass power Py0, such as the preset pass power Py0 can be set to a size range of 1.5MW-2.5MW.

[0054] Specifically, in the power prediction module, when judging the predicted power of the wind turbine, the real-time monitoring and judgment of the operating status of the wind turbine is achieved through accurate calculation and comparison of the predicted power, thereby improving the operating efficiency and safety of the wind turbine.

[0055] Specifically, when the power prediction module optimizes and adjusts the predicted power Py, the standard root mean square error value R is calculated according to the wind power operation power and the power prediction value, and R= ,in: is the actual value of wind power operation power, is the power prediction value, N is the actual total number of samples, and i is the actual wind power operation rate; The power prediction module compares the standard root mean square error value R with the preset standard root mean square error value R0, judges the error range of the predicted power Py according to the comparison result, and optimizes and adjusts the predicted power Py according to the judgment result, wherein: When R>R0, the power prediction module determines that the error range of the predicted power Py is unacceptable and does not pass the power prediction, optimizes and adjusts the predicted power Py, and sets the error adjustment coefficient β to optimize and adjust the predicted power Py, setting β=e -[(R-R0)+0.7] +0.15, e is the base of the natural logarithm, and the predicted power after optimization and adjustment is Py2=β×Py; When R≤R0, the power prediction module determines that the error range of the predicted power Py is within an acceptable range and passes the power prediction without optimizing and adjusting the predicted power Py.

[0056] Specifically, the standard root mean square error value R refers to a statistic that measures the difference between the predicted value and the actual value, the preset standard root mean square error value R0 refers to a threshold value set to determine whether the prediction accuracy is qualified, and this embodiment does not preset the standard root mean square error value R0, the error adjustment coefficient β refers to a coefficient used to adjust the predicted power error range, and the optimized adjusted predicted power Py2 refers to the predicted power value adjusted by the error adjustment coefficient β; Specifically, in the power prediction module, when optimizing and adjusting the wind power prediction, the root mean square error is calculated and compared with the preset standard, and the predicted power is optimized and adjusted according to the comparison result, so as to accurately judge the error range of the predicted power and improve the accuracy and efficiency of the wind power prediction.

[0057] Specifically, when the decision execution module integrates and analyzes the analysis and calculation data, it inputs the analysis and calculation data into the intelligent sensor for integration, and analyzes the integrated analysis and calculation data through the fault diagnosis model to obtain a status detection signal.

[0058] Specifically, the smart sensor refers to a sensor with data processing, network communication and self-diagnosis capabilities. This embodiment does not limit the type of smart sensors. Relevant technical personnel in this field can freely set them according to actual conditions, as long as the need to integrate the operation and analysis data is met. For example, the type of smart sensor can be set to an integrated meteorological sensor. The fault diagnosis model refers to a neural network model that obtains a state detection signal after integrating and analyzing the analysis operation data. This embodiment sets the basic framework of the fault diagnosis model to a convolutional neural network model, and uses historical analysis operation data-historical state detection signals as a diagnostic model construction data set, and uses 75% of the diagnostic model construction data set as a diagnostic model training set. 25% of the diagnostic model construction data set is used as the diagnostic model verification set. The convolutional neural network model is trained according to the diagnostic model training set to obtain the trained convolutional neural network model. The trained convolutional neural network model is verified according to the diagnostic model verification set. After the accuracy reaches 97%, the trained convolutional neural network model is output as a fault diagnosis model. This embodiment does not specifically limit the construction parameters of the convolutional neural network model. Those skilled in the art can freely set them according to actual conditions, as long as the requirements of state detection are met. For example, the loss function of the convolutional neural network model can be set to a cross entropy function. The state detection signal refers to a signal reflecting the operating status and fault information of the wind turbine.

[0059] Specifically, in the decision-making execution module, when the analytical calculation data is integrated and analyzed, the application of intelligent sensors and fault diagnosis models can achieve accurate monitoring and fault diagnosis of the operating status of the wind turbine, thereby improving the operating efficiency, operation and maintenance management level and economic benefits of the wind turbine.

[0060] Specifically, when the decision execution module performs decision execution on the state detection signal, the state detection signal is simulated and exercised through the signal simulation exercise model, and the predicted power is executed according to the exercise result, and a signal simulation exercise model is constructed, wherein the simulation exercise database is divided into a 70% exercise training set, a 20% exercise verification set and a 10% exercise test set, the exercise training set is input into the decision tree model to train the decision tree model, and the exercise verification set is input into the trained decision tree model, and the trained decision tree model is tested. The decision tree model is iteratively optimized for hyperparameters, and the drill test set is input into the iteratively optimized decision tree model to perform a drill test on the iteratively optimized decision tree model to obtain the drill test results. The total number of samples in the drill test set is set to Y0, the number of correct drill test samples is set to Y1, and the drill test accuracy is set to Y, Y=Y1 / Y0. The drill test accuracy Y is compared with the preset drill test accuracy Y0. According to the comparison results, the training compliance of the iteratively optimized decision tree model is judged, and the judgment results are output, where: When Y≥Y0, it is determined that the iteratively optimized decision tree model is successfully trained, a decision is executed on the predicted power, and the iteratively optimized decision tree model is output as a signal simulation exercise model; When Y<Y0, it is determined that the training of the iteratively optimized decision tree model has failed, and the simulation exercise database is updated to obtain an updated simulation exercise database. The decision tree model is trained, hyperparameters are iteratively optimized, and exercised and tested according to the updated simulation exercise database until the decision tree model is successfully trained.

[0061] Specifically, the decision execution refers to making decisions and judging the state detection signals of wind turbines according to the prediction results of the decision tree model, and performing corresponding operations. The signal simulation drill model refers to a model constructed based on the data and algorithms of analytical operations for simulating drills and decision analysis of state detection signals. The simulation drill database refers to a data set used for simulation drills. The drill training set refers to data divided from the simulation drill database. The drill verification set refers to the remaining data divided from the simulation drill database. The drill test set refers to a data set divided from the simulation drill database and independent of the training set and the verification set. The decision tree model refers to a machine learning model that makes decisions based on a tree structure. The hyperparameter iterative optimization refers to adjusting and optimizing the hyperparameters in the decision tree model. The drill test refers to the process of using the drill test set to test the trained and optimized decision tree model. The drill test result The result refers to the model performance evaluation result obtained after the drill test, the total number of samples in the drill test set refers to the total number of samples in the drill test set, the correct number of drill test samples refers to the number of samples correctly predicted by the model in the drill test, the drill test accuracy refers to the indicator for evaluating the performance of the decision tree model through the drill test set, the preset drill test accuracy refers to the set expected drill test accuracy used to determine whether the decision tree model has been successfully trained, the training compliance status refers to whether the decision tree model meets the preset performance requirements after training and optimization, and the update refers to modifying and supplementing the data and functions in the simulation drill database. This embodiment does not limit the method for updating the simulation drill database. Relevant technical personnel in this field can freely set it according to actual conditions, and only need to meet the needs of the updated simulation drill database for training the decision tree model, iterative optimization of hyperparameters and drill testing. For example, the update method can be set to model update.

[0062] Specifically, in the decision execution module, when making decisions on the status detection signal, a decision tree model is constructed through a signal simulation exercise model to divide the data set, train the model, iteratively optimize the hyperparameters, and conduct exercise tests. The optimized decision tree model makes decisions on the predicted power, realizes decision simulation exercises on the status detection signal, and trains and optimizes the decision tree model, thereby improving the accuracy and efficiency of decision execution as well as the operating efficiency and management level of the entire system.

[0063] See also Figure 2 As shown, it is a flow chart of the new energy station operation optimization method based on smart sensors in this embodiment, and the method includes: Step S1, collecting wind turbine operation data and environmental data; Step S2, performing data preprocessing on the wind turbine operating data and the environmental data according to the data preprocessing method to obtain preprocessed wind turbine operating data and preprocessed environmental data; Step S3, calculating the generator operation state coefficient according to the wind speed data in the preprocessed wind power generation operation data, and performing fault detection and maintenance on the generator according to the generator operation state coefficient; Step S4, calculating the environmental state parameters according to the preprocessed environmental data, judging the state of the preprocessed environmental data according to the environmental state parameters, and optimizing the generator operation state coefficient when the state of the preprocessed environmental data is abnormal; Step S5, judging the effectiveness of manual intervention according to the ambient temperature in the preprocessed environmental data, and optimizing the environmental state parameters according to the result of judging the effectiveness of manual intervention; Step S6, using the pre-processed wind power generation operation data, the generator operation state and the optimized environmental state parameters as analysis operation data, and transmitting the analysis operation data; Step S7, performing power prediction on the analyzed operation data to obtain predicted power, which is also used to detect and repair the generator according to the predicted power; Step S8, calculating a standard root mean square error value according to the wind power operation power and the power prediction value, and optimizing and adjusting the predicted power according to the standard root mean square error value; Step S9, integrating and analyzing the analysis operation data according to the intelligent sensor to obtain a state detection signal; Step S10, performing a decision simulation exercise on the state detection signal, and executing a decision on the predicted power according to the decision simulation exercise result. So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A new energy station operation optimization system based on intelligent sensors, characterized in that: The system comprises: A data collection module is used to collect wind turbine operation data and environmental data; A data preprocessing module, used to perform data preprocessing on the wind turbine operation data and environmental data according to a data preprocessing method to obtain preprocessed wind turbine operation data and preprocessed environmental data; An analysis and calculation module, used to calculate the generator operation state coefficient according to the wind speed data in the preprocessed wind power generation operation data, and to perform fault detection and repair on the generator according to the generator operation state coefficient, and also used to calculate the environmental state parameters according to the preprocessed environmental data, and to judge the state of the preprocessed environmental data according to the environmental state parameters, and to optimize the generator operation state coefficient when the state of the preprocessed environmental data is abnormal, and also used to judge the effectiveness of manual intervention according to the ambient temperature in the preprocessed environmental data, and to optimize the environmental state parameters according to the result of the judgment of the effectiveness of manual intervention; A data transmission module, used to use the pre-processed wind power generation operation data, the generator operation state and the optimized environmental state parameters as analysis operation data, and transmit the analysis operation data; A power prediction module is used to perform power prediction on the analysis and operation data to obtain the predicted power, and is also used to detect and repair the generator according to the predicted power, and is also used to calculate the standard root mean square error value according to the wind power operation power and the power prediction value, and optimize and adjust the predicted power according to the standard root mean square error value; The decision execution module is used to integrate and analyze the analytical operation data according to the intelligent sensor to obtain the status detection signal, and is also used to perform decision simulation exercises on the status detection signal, and to execute the decision on the predicted power according to the decision simulation exercise results.

2. The new energy station operation optimization system based on intelligent sensors according to claim 1 is characterized in that: When analyzing the operating status of the generator, the analysis and calculation module calculates the generator operating status coefficient S according to the wind speed data V in the preprocessed wind power generation operation data, the implementation power P corresponding to the wind speed in the selected time period, the wind speed data point i in the time period and the standard power Pr corresponding to the wind speed data point, and sets , where n is the threshold of wind speed data point i.

3. The new energy station operation optimization system based on intelligent sensors according to claim 2 is characterized in that: When the analysis and calculation module determines the type of the generator operating state, the generator operating state coefficient S is compared with a preset generator operating state coefficient standard value S0, the type of the generator operating state is determined according to the comparison result, and the generator is fault detected and repaired according to the determination result, wherein: When S≥S0, the generator operation status is determined to be normal, and no fault detection or maintenance is performed on the generator; When S<S0, the type of the generator operation state is determined to be abnormal, and the generator is fault detected and repaired.

4. The new energy station operation optimization system based on intelligent sensors according to claim 3 is characterized in that: When optimizing the operating state of the generator, the analysis and calculation module calculates the environmental state parameter E according to the pre-processed ambient temperature A, ambient humidity B and ambient pressure C, sets E=0.4A+0.3B+0.3C, compares the environmental state parameter E with the preset environmental state parameter standard value E0, judges the state of the environmental data according to the comparison result, and optimizes the generator operating state coefficient S according to the judgment result, wherein: When E≥E0, the analysis and operation module determines that the state of the preprocessed environmental data is normal, and does not optimize the generator operation state coefficient S; When E<E0, the analysis and operation module determines that the state of the pre-processed environmental data is abnormal, optimizes the generator operation state coefficient S, and calculates the optimization coefficient α, setting α=0.6×e -0.02×E +0.4×(0.002×E+0.65） , e is a natural number base, the generator operating state coefficient S is optimized according to the optimization coefficient α, the optimized generator operating state coefficient is Sα, and Sα=α×S is set.

5. The new energy station operation optimization system based on intelligent sensors according to claim 4 is characterized in that: When optimizing the environmental state parameter E, the analysis and operation module compares the ambient temperature A in the preprocessed environmental data with the preset ambient temperature AO, judges the effectiveness of the manual intervention according to the comparison result, and optimizes the environmental state parameter E according to the judgment result, wherein: When A≤A0, the effectiveness of the manual intervention is determined to be passed, and the analysis and calculation module does not optimize the environmental state parameter E; When A>A0, the effectiveness of the manual intervention is judged as not passed, and the analysis and calculation module optimizes the environmental state parameter E and calculates the environmental impact coefficient θ, setting θ=0.66+e -0.1×[(E0-E)+12]）] , e is a natural number base, the environmental state parameter E is optimized according to the environmental impact coefficient θ, the optimized environmental state parameter is Ee, and Ee=θ×E is set.

6. The new energy station operation optimization system based on intelligent sensors according to claim 1 is characterized in that: When the power prediction module performs power prediction on the analysis and operation data, the analysis and operation data is input into the power prediction model to construct a power prediction model, wherein the power prediction database is divided into a prediction training set of 75%, a prediction verification set of 15% and a prediction test set of 10%, the prediction training set is input into the decision tree structure model to train the decision tree structure model, and the prediction verification set is input into the trained decision tree structure model, the hyperparameters of the decision tree structure model are iteratively optimized, and the prediction test set is input into the iteratively optimized decision tree structure model to perform power test on the iteratively optimized decision tree structure model to obtain predicted power, the total number of predicted power samples is set to g0, the number of correctly predicted test samples is set to g, the predicted power accuracy is set to G, G=g / g0, the predicted power accuracy G is compared with the preset predicted power accuracy G0, the training compliance of the iteratively optimized decision tree structure model is judged according to the comparison result, and the judgment result is output, wherein: When G≥G0, it is determined that the iteratively optimized decision tree structure model training has reached the standard, and the iteratively optimized decision tree structure model is output as a power prediction model; When G<G0, it is determined that the training of the iteratively optimized decision tree structure model does not meet the standards, the power prediction database is updated to obtain an updated power prediction database, and the decision tree model is trained, hyperparameters are iteratively optimized and power tested according to the updated power prediction database until the training of the decision tree structure model meets the standards.

7. The new energy station operation optimization system based on intelligent sensors according to claim 6 is characterized in that: When judging the predicted power, the power prediction module calculates the predicted power Py according to the radius r, wind speed v, air density ρ and efficiency factor n of the wind turbine, and sets Py= , and compare the predicted power Py with the preset power Py0, and judge the predicted power Py according to the comparison result, where: When Py≥Py0, the power prediction module determines that the predicted power Py passes the power prediction and can continue to run; When Py<Py0, the power prediction module determines that the predicted power Py does not pass the power prediction and cannot continue to operate, and needs to be tested and repaired again.

8. The new energy station operation optimization system based on intelligent sensors according to claim 7 is characterized in that: When the power prediction module optimizes and adjusts the predicted power Py, it calculates the standard root mean square error value R according to the wind power operation power and the power prediction value, and sets R= ,in: is the actual value of wind power operation power, is the power prediction value, N is the actual total number of samples, and i is the actual wind power operation rate; The power prediction module compares the standard root mean square error value R with the preset standard root mean square error value R0, judges the error range of the predicted power Py according to the comparison result, and optimizes and adjusts the predicted power Py according to the judgment result, wherein: When R>R0, the power prediction module determines that the error range of the predicted power Py is unacceptable and does not pass the power prediction, optimizes and adjusts the predicted power Py, and sets the error adjustment coefficient β to optimize and adjust the predicted power Py, setting β=e -[(R-R0)+0.7] +0.15, e is the base of the natural logarithm, and the predicted power after optimization and adjustment is Py2=β×Py; When R≤R0, the power prediction module determines that the error range of the predicted power Py is within an acceptable range and passes the power prediction without optimizing and adjusting the predicted power Py.

9. The new energy station operation optimization system based on intelligent sensors according to claim 1 is characterized in that: When the decision execution module performs decision execution on the state detection signal, the state detection signal is simulated and exercised through the signal simulation exercise model, and the predicted power is executed according to the exercise result, and a signal simulation exercise model is constructed, wherein the simulation exercise database is divided into a 70% exercise training set, a 20% exercise verification set and a 10% exercise test set, the exercise training set is input into the decision tree model to train the decision tree model, and the exercise verification set is input into the trained decision tree model, and the trained decision tree model is tested. The hyperparameters of the tree model are iteratively optimized, and the training test set is input into the iteratively optimized decision tree model to perform a training test on the iteratively optimized decision tree model to obtain the training test results. The total number of samples in the training test set is set to Y0, the number of correct training test samples is set to Y1, and the training test accuracy is set to Y, Y=Y1 / Y0. The training test accuracy Y is compared with the preset training test accuracy Y0. According to the comparison results, the training compliance of the iteratively optimized decision tree model is judged, and the judgment results are output, where: When Y≥Y0, it is determined that the iteratively optimized decision tree model is successfully trained, a decision is executed on the predicted power, and the iteratively optimized decision tree model is output as a signal simulation exercise model; When Y<Y0, it is determined that the training of the iteratively optimized decision tree model has failed, and the simulation exercise database is updated to obtain an updated simulation exercise database. The decision tree model is trained, hyperparameters are iteratively optimized, and exercised and tested according to the updated simulation exercise database until the decision tree model is successfully trained.

10. A method applied to the new energy station operation optimization system based on intelligent sensors as described in any one of claims 1 to 9, characterized in that: The method comprises: Step S1, collecting wind turbine operation data and environmental data; Step S2, performing data preprocessing on the wind turbine operating data and the environmental data according to the data preprocessing method to obtain preprocessed wind turbine operating data and preprocessed environmental data; Step S3, calculating the generator operation state coefficient according to the wind speed data in the preprocessed wind power generation operation data, and performing fault detection and maintenance on the generator according to the generator operation state coefficient; Step S4, calculating the environmental state parameters according to the preprocessed environmental data, judging the state of the preprocessed environmental data according to the environmental state parameters, and optimizing the generator operation state coefficient when the state of the preprocessed environmental data is abnormal; Step S5, judging the effectiveness of manual intervention according to the ambient temperature in the preprocessed environmental data, and optimizing the environmental state parameters according to the result of judging the effectiveness of manual intervention; Step S6, using the pre-processed wind power generation operation data, the generator operation state and the optimized environmental state parameters as analysis operation data, and transmitting the analysis operation data; Step S7, performing power prediction on the analyzed operation data to obtain predicted power, which is also used to detect and repair the generator according to the predicted power; Step S8, calculating a standard root mean square error value according to the wind power operation power and the power prediction value, and optimizing and adjusting the predicted power according to the standard root mean square error value; Step S9, integrating and analyzing the analysis and calculation data according to the intelligent sensor to obtain a state detection signal; Step S10, performing a decision simulation exercise on the state detection signal, and executing a decision on the predicted power according to the decision simulation exercise result.

Images