Data control system and method based on artificial intelligence

Abstract

The invention discloses a data control system and method based on artificial intelligence, and particularly relates to the technical field of data control. Comprising a power equipment auxiliary equipment data control information base, a power equipment auxiliary equipment data acquisition module, a power equipment auxiliary equipment data feature extraction module, a power equipment auxiliary equipment data feature analysis module, a power equipment auxiliary equipment intelligent control decision module and a data control intelligent feedback module. The power equipment auxiliary equipment data acquisition module is used for arranging a sensor group at a key part based on a power equipment auxiliary equipment data control information base, and acquiring a first target parameter required by power equipment auxiliary equipment data control in real time according to a set sampling frequency; according to the method, the support vector machine is constructed through the mode recognition technology, the LSTM prediction model is constructed through the trend prediction technology, the extracted features are deeply analyzed, and the precision of data control is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data control, and specifically provides a data control system and method based on artificial intelligence. Background Art

[0002] In the management of auxiliary equipment of power equipment, data control is the core link to ensure the safe, stable operation and efficient maintenance of the equipment; the introduction of artificial intelligence technology has significantly improved the accuracy, efficiency and decision-making ability of data control for auxiliary equipment of power equipment through data-driven intelligent means.

[0003] Traditional data control methods mainly rely on hardware logic control, manual experience judgment and simple rule algorithms; hardware logic control builds logic circuits through hardware devices to realize functions such as equipment start-stop and protection; operators manually adjust equipment parameters according to historical experience; based on preset data rules, equipment control is performed to realize the control of power equipment through data.

[0004] Traditional data control methods have played an important role in a specific historical stage. However, with the complexity of the power system and the increasing demand for intelligence, the current data control of auxiliary equipment of power equipment faces many challenges. For example, manual adjustment has hysteresis and low efficiency; simple rule algorithms cannot effectively extract key features from a large amount of data of auxiliary equipment of power equipment and are difficult to deeply explore the complex laws and potential trends hidden behind the data of auxiliary equipment; the current control decisions are mainly based on preset fixed rules, lacking the adaptive ability to real-time data changes and complex working conditions, and the control accuracy is insufficient; the existing feedback adjustment mechanism has a slow response speed and a single adjustment method, and cannot be optimized according to the dynamic characteristics of the equipment. Therefore, there are many problems in the intelligence aspects such as data analysis, control decision-making and feedback adjustment in the current data control of auxiliary equipment of power equipment, which seriously restrict the efficient and stable operation of auxiliary equipment of power equipment; therefore, developing an advanced data control system and method based on artificial intelligence is crucial to ensure that the power system always maintains the best operating state. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a data control system and method based on artificial intelligence to solve the problems of lack of adaptive ability, low efficiency, insufficient control accuracy, etc. proposed in the above background art.

[0006] To achieve the above object, the present invention provides the following technical solution: A data control system based on artificial intelligence, comprising: Power Equipment Auxiliary Equipment Data Control Information Base: Through big data technology, it stores and updates in real time the standard parameters required for the data control of the target power equipment auxiliary equipment, and constructs a power equipment auxiliary equipment data control information base. The standard parameters include the operation design data, required environmental data, and fault data of the target power equipment auxiliary equipment; Power Equipment Auxiliary Equipment Data Acquisition Module: Based on the power equipment auxiliary equipment data control information base, sensor groups are arranged at key positions, and the first target parameters required for the data control of the power equipment auxiliary equipment are collected in real time according to the set sampling frequency and transmitted to the power equipment auxiliary equipment data feature extraction module; Power Equipment Auxiliary Equipment Data Feature Extraction Module: Through big data technology, feature extraction is performed on the first target parameters obtained by the data acquisition module to obtain the second target parameters required for the data control of the power equipment auxiliary equipment, and they are transmitted to the power equipment auxiliary equipment data feature analysis module. The first target parameters include the first operation state parameters, the first required environmental parameters, and the first fault parameters; Power Equipment Auxiliary Equipment Data Feature Analysis Module: Through pattern recognition technology and trend prediction technology, feature analysis is performed on the second target parameters obtained by the data feature extraction module to obtain a feature analysis result, which is transmitted to the power equipment auxiliary equipment intelligent control decision-making module. The feature analysis result includes a pattern recognition result and a trend prediction result; Power Equipment Auxiliary Equipment Intelligent Control Decision-making Module: Generates corresponding control strategies according to the feature analysis results obtained by the feature analysis module, and transmits the results after the execution of the control strategies to the data control intelligent feedback module; Data Control Intelligent Feedback Module: Used to monitor the second target parameter results after the execution of the control strategies, and feedback to the power equipment auxiliary equipment intelligent control decision-making module according to the abnormal monitoring results.

[0007] Preferably, an artificial intelligence-based data control method includes the following steps: Step 1: Through big data technology, store and update in real time the standard parameters required for the data control of the target power equipment auxiliary equipment, and construct a power equipment auxiliary equipment data control information base; Step 2: According to the power equipment auxiliary equipment data control information base in Step 1, arrange sensor groups at key positions, and collect the first target parameters required for the data control of the power equipment auxiliary equipment in real time according to the set sampling frequency; Step 3: Through big data technology, perform feature extraction on the first target parameters in Step 2 to obtain the second target parameters required for the data control of the power equipment auxiliary equipment; Step 4: Through pattern recognition technology and trend prediction technology, perform feature analysis on the second target parameters in Step 3 to obtain a feature analysis result; Step 5: Generate corresponding control strategies based on the feature analysis results of Step 4; Step 6: Monitor the results of the second target parameter after the execution of the control strategy in Step 5, and feedback to Step 5 according to the abnormal monitoring results.

[0008] Technical effects and advantages of the present invention: 1. In the present invention, a sensor group is arranged at key positions through the data acquisition module to collect data in real time and ensure the timeliness and accuracy of the data; through big data analysis technology, multi-dimensional feature extraction is performed on the equipment operation status, environment and fault data. These refined feature extraction methods can accurately capture the internal laws and abnormal changes of the data, which helps to improve the data control efficiency; 2. In the present invention, a support vector machine is constructed through pattern recognition technology and an LSTM prediction model is constructed through trend prediction technology to deeply analyze the extracted features, effectively improving the accuracy of data control, being able to more accurately identify the normal operation mode and abnormal mode of the equipment, and thus improving the adaptive ability of data control; 3. In the present invention, the intelligent control decision module automatically generates corresponding control strategies, such as shutdown, parameter adjustment, parameter warning, etc., according to the results of the feature analysis module without manual intervention, realizing the intelligence of control; through the abnormal severity score and trend prediction risk score, different levels of abnormalities and potential risks are classified and processed, improving the pertinence and effectiveness of control decisions. Brief Description of the Drawings

[0009] Figure 1 It is a schematic diagram of the overall process of the present invention.

[0010] Figure 2 It is a schematic diagram of the method process of the present invention. Detailed Embodiments

[0011] The following will combine the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention. Embodiment 1

[0012] Please refer to Figure 1As shown in the figure, the present invention provides an artificial intelligence-based data control system, including a power equipment auxiliary equipment data control information library, a power equipment auxiliary equipment data acquisition module, a power equipment auxiliary equipment data feature extraction module, a power equipment auxiliary equipment data feature analysis module, a power equipment auxiliary equipment intelligent control decision-making module, and a data control intelligent feedback module; It should be specifically noted in this embodiment that the present invention focuses on controlling the data of power equipment auxiliary equipment. Examples of power equipment auxiliary equipment include power cable installation traction equipment, power tower de-icing equipment, etc. The core functions of the power system are the production, transmission, distribution, and use of electric energy. Equipment such as generators, transformers, and power cables directly participate in this core electric energy processing process. The main function of the power cable traction equipment is to assist in the laying and installation of power cables and does not directly participate in the core links such as the production, transmission, distribution, or use of electric energy. It only provides support for ensuring the correct placement of this key electric energy transmission component, the power cable, and belongs to the equipment that assists in realizing the core functions of the power system. Therefore, it belongs to power equipment auxiliary equipment.

[0013] The power equipment auxiliary equipment data control information library is connected to the power equipment auxiliary equipment data acquisition module. The power equipment auxiliary equipment data feature extraction module is respectively connected to the power equipment auxiliary equipment data acquisition module and the power equipment auxiliary equipment data feature analysis module. The power equipment auxiliary equipment intelligent control decision-making module is respectively connected to the power equipment auxiliary equipment data feature analysis module and the data control intelligent feedback module.

[0014] Power equipment auxiliary equipment data control information library: Through big data technology, it stores and updates in real time the standard parameters required for the data control of target power equipment auxiliary equipment, and constructs a power equipment auxiliary equipment data control information library. The standard parameters include the operation design data, required environmental data, and fault data of the target power equipment auxiliary equipment; It should be specifically noted in this embodiment that the operation design data of power equipment auxiliary equipment are, for example, voltage, current, power, speed, etc.; the environmental data are, for example, temperature, humidity, dust concentration, etc.; the fault data are, for example, fault codes, fault occurrence times, etc. The power equipment auxiliary equipment data control information library can be obtained through public data sources (such as State Grid bidding parameters) or enterprise-owned data.

[0015] Power equipment auxiliary equipment data acquisition module: Based on the power equipment auxiliary equipment data control information library, sensor groups are arranged at key positions, and the first target parameters required for the data control of power equipment auxiliary equipment are collected in real time according to the set sampling frequency and transmitted to the power equipment auxiliary equipment data feature extraction module; In this embodiment, it should be specifically noted that, for example, the auxiliary equipment of the target power equipment is a traction equipment for power cable installation; the sensor group includes a temperature sensor, a current sensor, a voltage sensor, a vibration sensor, an ambient humidity sensor, etc.; tag information such as a timestamp and a location identifier has been added to each collected data point, where the timestamp is used to record the specific time of data collection, and the location identifier is used to clarify the source location of data collection (such as specific sensor numbers, equipment parts, etc.) for subsequent accurate analysis and positioning of the data; the collected raw data has been preliminarily processed, such as filtering, denoising, normalization, etc., to improve the quality and usability of the data.

[0016] Data feature extraction module for auxiliary equipment of power equipment: Through big data technology, extract features from the first target parameters obtained by the data collection module to obtain the second target parameters required for the data control of the auxiliary equipment of the power equipment, and transmit them to the data feature analysis module for the auxiliary equipment of the power equipment. The first target parameters include the first operating state parameters, the first required environmental parameters, and the first fault parameters. Data feature analysis module for auxiliary equipment of power equipment: Through pattern recognition technology and trend prediction technology, perform feature analysis on the second target parameters obtained by the data feature extraction module to obtain a feature analysis result, and transmit it to the intelligent control decision-making module for the auxiliary equipment of the power equipment. The feature analysis result includes a pattern recognition result and a trend prediction result. Intelligent control decision-making module for auxiliary equipment of power equipment: Generate corresponding control strategies according to the feature analysis results obtained by the feature analysis module, and transmit the results after the execution of the control strategies to the intelligent feedback module for data control. Intelligent feedback module for data control: Used to monitor the results of the second target parameters after the execution of the control strategies, and feedback to the intelligent control decision-making module for the auxiliary equipment of the power equipment according to the abnormal monitoring results. In this embodiment, it should be specifically noted that the data feature extraction module for the auxiliary equipment of the power equipment includes a second target parameter feature extraction unit, and the second target parameter feature extraction unit includes a second operating state parameter extraction unit, a second required environmental parameter extraction unit, and a second fault parameter extraction unit; among them, the second operating state parameter extraction unit is used to extract features from the first operating state parameters, the second required environmental parameter extraction unit is used to extract features from the first required environmental parameters, the second fault parameter extraction unit is used to extract features from the first fault parameters, and the extraction of the second target parameters includes the following steps: Step 1: Second operating state parameter extraction unit: During the data control cycle, based on the types of operating design data in the data control information library for the auxiliary equipment of the power equipment, obtain the first operating state parameter RS for the data control of the target auxiliary equipment of the power equipment. , rs irepresents the value of the i-th operating state parameter, and n represents the number of types of operating state parameters; then, each operating state parameter is sampled m times to obtain the mean μ(rs i ) and the standard deviation σ(rs i ); finally, through big data analysis technology, the second operating state parameter OSCV of the target power equipment auxiliary equipment is obtained, , , cv i represents the eigenvalue of the i-th operating state parameter, and n also represents the number of types of eigenvalues of the operating state parameters; It should be specifically noted in this embodiment that, for example, when the auxiliary equipment is a traction device for installing power cables, the set of operating state parameters includes the traction speed v(t), the motor current I(t), the hydraulic pressure P(t), the gearbox temperature T g (t), etc.

[0017] Step 2: Second required environmental parameter extraction unit: During the data control cycle, based on the required environmental data in the data control information library of the power equipment auxiliary equipment, the first required environmental parameter ED for the data control of the target power equipment auxiliary equipment is obtained, , e j represents the value of the j-th environmental parameter, and k represents the number of types of environmental parameters; then, the value e of each environmental parameter j is compared with the allowed value (e j )0 of each environmental parameter to obtain the coefficient η(Δe j ) by which each environmental parameter exceeds the allowed value, ; finally, through big data analysis technology, the second required environmental parameter ECVD of the target power equipment auxiliary equipment is obtained, , vd j represents the eigenvalue of the j-th environmental parameter, and k also represents the number of types of eigenvalues of the environmental parameters, vd j =η(Δe j ); It should be specifically noted in this embodiment that, for example, when the auxiliary equipment is a traction device for installing power cables, the environmental data set includes the environmental temperature T env (t), the wind speed w(t), and the ground slope θ(t).

[0018] Step 3: Second fault parameter extraction unit: During the data control cycle, based on the fault data in the data control information library of the power equipment auxiliary equipment, the first fault parameter for the data control of the target power equipment auxiliary equipment is obtained. The first fault parameter includes the number of fault types I and the number of fault occurrences n_ff, the number of warning types N and the number of warning occurrences n_wf; finally, through big data analysis technology, the second fault parameter FCVD of the target power equipment auxiliary equipment is obtained, , η(n_ff) represents the growth rate dataset of the number of I types of fault types, , , (n_ff) I represents the growth rate of the number of the I-th type of fault, (n_ff) I (T) and (n_ff) I (T-3) respectively represent the number of faults in the current data control cycle and the number of faults in the third cycle closest to the history of the I-th type of fault. η(n_wf) represents the growth rate dataset of the number of N types of warning types, , , (n_wf) N represents the growth rate of the number of the I-th type of warning, (n_wf) N (T) and (n_wf) N (T-3) respectively represent the number of warnings in the current data control cycle and the number of warnings in the third cycle closest to the history. If η(n_ff) I ≤0, then η(n_ff) I =0, otherwise it is the value of η(n_ff) I . If η(n_wf) N ≤0, then η(n_wf) N =0, otherwise it is the value of η(n_wf) N ; In this embodiment, it should be specifically noted that the fault types and numbers of the high-voltage power cable traction machine are 2 broken strands, and the warning types and numbers are 1 cable slip alarm, etc.

[0019] In this embodiment, it should be specifically noted that the power equipment auxiliary equipment data feature analysis module includes a pattern recognition unit and a trend prediction unit. The feature analysis includes the following steps: Step 1: Pattern recognition unit: First, a support vector machine model is constructed through pattern recognition technology. Its input layer receives the second target historical parameters and corresponding operation mode category labels of each data control cycle, and outputs the trained model; then the second target parameters are input into the support vector machine model for pattern recognition to obtain the operation mode category of the target power equipment auxiliary equipment. The operation mode category includes normal and abnormal. The pattern recognition includes the following steps: Step 1.1: First, the second target parameters are input into the support vector machine model for pattern recognition. The second target parameters include the second operating state parameter OSCV, the second required environmental parameter ECVD, and the second fault parameter FCVD, to obtain the second target parameter feature vector X, X = [OSCV, ECVD, FCVD]; then the support vector machine decision function F(X) is obtained, , where sgn represents the sign function, outputting +1 and -1, indicating normal and abnormal respectively, and α J represents the Lagrange multiplier, and α J >0. The corresponding eigenvector is the support vector, and y J represents the label of the J-th support vector, that is, +1 and -1, and X J represents the eigenvalue of the J-th support vector. M represents the number of types of eigenvalues. M = n + k + 2, where n represents the number of types of eigenvalues of the operating state parameters, k represents the number of types of eigenvalues of the environmental parameters, and 2 represents the two eigenvalues in the second fault parameter FCVD. K(X J , X) represents the kernel function, and K(X J , X) = exp(-0.1×||X J - X|| 2 ). b represents the bias term, that is, the intercept of the decision hyperplane, and ||·|| represents the Euclidean distance between X J and X, which is obtained through optimization calculation; finally, the pattern recognition result is obtained, including normal and abnormal; Step 2: Trend prediction unit: First, construct an LSTM prediction model through trend prediction technology. Its input layer receives the second target historical parameters of each data control cycle according to the time series, and the output layer predicts the second target historical parameters of the next L data control cycles. The LSTM prediction model is , where T represents the data control cycle, and F(t + L×T) represents the second target parameters of the next L data control cycles, including the second operating state parameter OSCV, the second required environmental parameter ECVD, and the second fault parameter FCVD; finally, the trend prediction result is obtained; It should be specifically noted in this embodiment that SVM is a supervised learning algorithm that realizes classification by constructing an optimal hyperplane and is suitable for high-dimensional and non-linear data (such as multi-source parameters of auxiliary equipment of power equipment); the long short-term memory network (LSTM) is a time series model that captures long-term dependencies through memory units and is suitable for predicting the future trends of parameters of auxiliary equipment of power equipment.

[0020] In this embodiment, it should be specifically noted that the intelligent control decision module of the auxiliary equipment of the power equipment generates corresponding control strategies according to the feature analysis results obtained by the feature analysis module, including the following steps: Step 1: First, analyze according to the result of pattern recognition being abnormal to obtain the abnormal severity score S F , and S F =(X J - X J 0 ) / X J 0 , where X J and X J0 respectively represent the eigenvalue of the J-th support vector and the corresponding threshold. If S F ≥0.8, then generate an auxiliary device shutdown control strategy corresponding to the anomaly; if 0.8 > S F ≥0.5, then generate a parameter control strategy corresponding to the anomaly; if S F <0.5, then generate a parameter warning control strategy corresponding to the anomaly; Step 2: Then analyze according to the result of trend prediction. If LSTM(X J ) ≤ X J 0 , obtain the trend prediction risk score RS1, RS1 = 1 - [|LSTM(X J ) - X J 0 | / X J 0 , where LSTM(X J ) represents LSTM(X J ) obtained through trend prediction. If RS1 ≥ 0.8, then generate a corresponding parameter prevention control strategy; if LSTM(X J ) > X J 0 , obtain the trend prediction risk score RS2, RS2 = LSTM(X J ) - X J 0 / X J 0 , if RS2 ≥ 0.8, then generate a corresponding auxiliary device shutdown prevention control strategy; if 0.8 > RS2 ≥ 0.5, then generate a corresponding parameter prevention control strategy; if RS2 < 0.5, then generate a corresponding parameter warning prevention control strategy; In this embodiment, it should be specifically noted that the data control intelligent feedback module is used to monitor the second target parameter result after the execution of the control strategy, and feedback it to the intelligent control decision module of the power equipment auxiliary device according to the anomaly monitoring result. The intelligent feedback includes the following steps: Step 1: Through big data technology, obtain the output value (X J ) out of the eigenvalue of the J-th support vector after being controlled by the intelligent control decision module, and compare it with the corresponding threshold X J 0 to obtain the deviation coefficient model η[(X J ) out of the eigenvalue of the J-th support vector, η[(X J ) out = (X J ) out - X J0 / X J 0 If η[(X J ) out >0, it indicates that the monitoring result is abnormal, and then feedback information of the eigenvalue of the Jth support vector is sent. Otherwise, it indicates that the intelligent control decision is normal and there is no feedback; Step 2: Based on Step 1, input the eigenvalues of the remaining M - 1 support vectors into the deviation coefficient model, and send feedback information according to the abnormal monitoring result, so as to ensure that the system always maintains the best operating state;

[0021] Embodiment 2: For the parts not described in detail in this embodiment, refer to the description content of Embodiment 1. A data control method based on artificial intelligence includes the following steps: Step 1: Through big data technology, store and update in real time the standard parameters required for data control of auxiliary equipment of target power equipment, and construct a data control information library for auxiliary equipment of power equipment; Step 2: According to the data control information library of the auxiliary equipment of the power equipment in Step 1, arrange a sensor group at key positions, and collect the first target parameters required for data control of the auxiliary equipment of the power equipment in real time according to the set sampling frequency; Step 3: Through big data technology, perform feature extraction on the first target parameters in Step 2 to obtain the second target parameters required for data control of the auxiliary equipment of the power equipment; Step 4: Through pattern recognition technology and trend prediction technology, perform feature analysis on the second target parameters in Step 3 to obtain a feature analysis result; Step 5: Generate corresponding control strategies according to the feature analysis result in Step 4; Step 6: Monitor the result of the second target parameters after the execution of the control strategy in Step 5, and feedback to Step 5 according to the abnormal monitoring result.

[0022] Secondly: In the attached drawings of the disclosed embodiments of the present invention, only the structures related to the disclosed embodiments are involved. For other structures, reference can be made to the general design. Without conflict, the same embodiment and different embodiments of the present invention can be combined with each other; Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A data control system based on artificial intelligence, characterized in that: include: Power equipment auxiliary equipment data control information library: through big data technology, the standard parameters required for data control of target power equipment auxiliary equipment are stored and updated in real time, and the power equipment auxiliary equipment data control information library is constructed. The standard parameters include the operation design data, required environmental data and fault data of the target power equipment auxiliary equipment; Power equipment auxiliary equipment data acquisition module: Based on the power equipment auxiliary equipment data control information library, a sensor group is arranged at key positions to collect the first target parameters required for power equipment auxiliary equipment data control in real time according to the set sampling frequency, and transmit them to the power equipment auxiliary equipment data feature extraction module; Power equipment auxiliary equipment data feature extraction module: through big data technology, the first target parameter obtained by the data acquisition module is feature extracted to obtain the second target parameter required for the power equipment auxiliary equipment data control, and transmitted to the power equipment auxiliary equipment data feature analysis module, the first target parameter includes the first operating state parameter, the first required environment parameter and the first fault parameter; Power equipment auxiliary equipment data feature analysis module: through pattern recognition technology and trend prediction technology, the second target parameter obtained by the data feature extraction module is subjected to feature analysis to obtain feature analysis results, which are transmitted to the power equipment auxiliary equipment intelligent control decision module. The feature analysis results include pattern recognition results and trend prediction results. Intelligent control decision module for power equipment auxiliary equipment: generates corresponding control strategies according to the characteristic analysis results obtained by the characteristic analysis module, and transmits the results after the control strategy is executed to the data control intelligent feedback module; Data control intelligent feedback module: used to monitor the second target parameter result after the execution of the control strategy, and feedback to the intelligent control decision module of the power equipment auxiliary equipment based on the abnormal monitoring results.

2. The data control system based on artificial intelligence according to claim 1, characterized in that: The power equipment auxiliary equipment data feature extraction module includes a second target parameter feature extraction unit, and the second target parameter feature extraction unit includes a second operating state parameter extraction unit, a second required environment parameter extraction unit, and a second fault parameter extraction unit; The second operating state parameter extraction unit is used to extract features from the first operating state parameter, the second required environment parameter extraction unit is used to extract features from the first required environment parameter, and the second fault parameter extraction unit is used to extract features from the first fault parameter.

3. The data control system based on artificial intelligence according to claim 2, characterized in that: The second operation state parameter extraction unit: within the data control period, based on the operation design data type in the power equipment auxiliary equipment data control information library, obtains the first operation state parameter RS ​​of the target power equipment auxiliary equipment data control, , rs i represents the value of the i-th operating state parameter, and n represents the number of types of operating state parameters; then each operating state parameter is sampled m times to obtain the mean value μ(rs i ) and standard deviation σ(rs i ); Finally, the second operating state parameter OSCV of the target power equipment auxiliary equipment is obtained through big data analysis technology. , cv i represents the characteristic value of the i-th operating state parameter, and n also represents the number of types of characteristic values ​​of the operating state parameter.

4. The data control system based on artificial intelligence according to claim 2, characterized in that: The second required environmental parameter extraction unit: within the data control period, based on the required environmental data in the power equipment auxiliary equipment data control information library, obtains the first required environmental parameter ED of the target power equipment auxiliary equipment data control, , e j represents the value of the jth environmental parameter, k represents the number of types of environmental parameters; then for each environmental parameter value e j The value allowed for each environmental parameter (e j )0, and the coefficient η(Δe j ); Finally, the second required environmental parameter ECVD of the target power equipment auxiliary equipment is obtained through big data analysis technology. , vd j represents the characteristic value of the jth environmental parameter, k also represents the number of characteristic values ​​of the environmental parameter, vd j =η(Δe j ).

5. The data control system based on artificial intelligence according to claim 2, characterized in that: The second fault parameter extraction unit: within the data control cycle, based on the fault data in the power equipment auxiliary equipment data control information library, obtains the first fault parameter of the target power equipment auxiliary equipment data control, the first fault parameter including the number of fault types I and the number of faults n_ff, the number of warning types N and the number of warnings n_wf; finally, through the big data analysis technology, the second fault parameter FCVD of the target power equipment auxiliary equipment is obtained, , η(n_ff) represents the growth rate data set of the number of fault types I, , (n_ff) I represents the growth rate of the number of the first type of fault, η(n_wf) represents the growth rate data set of the number of N types of warning types, , (n_wf) N represents the growth rate of the number of the first type of warning, if η(n_ff) I ≤0, then η(n_ff) I =0, otherwise it is η(n_ff) I If η(n_wf) N ≤0, then η(n_wf) N =0, otherwise η(n_wf) N The numerical value of .

6. The data control system based on artificial intelligence according to claim 1, characterized in that: The power equipment auxiliary equipment data feature analysis module includes a pattern recognition unit and a trend prediction unit: The pattern recognition unit: firstly inputs the second target parameter into the support vector machine model for pattern recognition, the second target parameter includes the second operating state parameter OSCV, the second required environment parameter ECVD and the second fault parameter FCVD, and obtains the second target parameter feature vector X, X=[OSCV, ECVD, FCVD]; then obtains the support vector machine decision function F(X), , sgn means compliance function, output +1 and -1, indicating normal and abnormal, respectively, α J represents the Lagrange multiplier, α J >0 is the corresponding eigenvector, y J represents the label of the Jth support vector, i.e., +1 and -1, X J represents the eigenvalue of the Jth support vector, M represents the number of eigenvalue types, M=n+k+2, n represents the number of eigenvalue types of the operating state parameter, k represents the number of eigenvalue types of the environmental parameter, 2 represents the two eigenvalues ​​in the second fault parameter FCVD, K(X J ,X) represents the kernel function, K(X J ,X)=exp(-0.1×||X J -X|| 2 ), b represents the bias term, i.e., the intercept of the decision hyperplane, and ||·|| represents X J and the Euclidean distance of X; finally, the pattern recognition results are obtained, including normal and abnormal; The trend prediction unit: firstly constructs an LSTM prediction model through trend prediction technology, wherein the input layer receives the second target historical parameters of each data control cycle according to the time series, and the output layer predicts the second target historical parameters of the next L data control cycles. The LSTM prediction model is , T represents the data control cycle, F(t+L×T) represents the second target parameters of the future L data control cycles, including the second operating state parameter OSCV, the second required environment parameter ECVD and the second fault parameter FCVD; finally, the trend prediction result is obtained.

7. The data control system based on artificial intelligence according to claim 1, characterized in that: The intelligent control decision module for auxiliary equipment of power equipment includes the following steps: Step 1: First, analyze the results of pattern recognition as abnormalities to obtain the abnormality severity score S F , S F =(X J -X J 0 ) / X J 0 , X J and X J 0 Represent the eigenvalue and corresponding threshold of the J-th support vector respectively. If S F ≥0.8, then generate the auxiliary equipment shutdown control strategy corresponding to the abnormality; if 0.8>S F ≥0.5, then generate the parameter control strategy corresponding to the abnormality; if S F <0.5, then a parameter warning control strategy corresponding to the abnormality is generated; Step 2: Then analyze the trend prediction results. If LSTM(X J )≤X J 0 , get the trend prediction risk score RS1, RS1=1-[|LSTM(X J )-X J 0 | / X J 0 ],LSTM(X J ) represents the LSTM(X J ), if RS1≥0.8, then generate the corresponding parameter prevention control strategy; if LSTM(X J )>X J 0 , get the trend prediction risk score RS2, RS2=LSTM(X J )-X J 0 / X J 0 If RS2≥0.8, the corresponding auxiliary equipment shutdown prevention control strategy is generated; if 0.8>RS2≥0.5, the corresponding parameter prevention control strategy is generated; if RS2<0.5, the corresponding parameter early warning prevention control strategy is generated.

8. The data control system based on artificial intelligence according to claim 1, characterized in that: The intelligent feedback in the data control intelligent feedback module includes the following steps: Step 1: Using big data technology, obtain the output value (X J ) out , and the corresponding threshold X J 0 By comparison, we can obtain the deviation coefficient model η[(X J ) out ],η[(X J ) out ]=(X J ) out -X J 0 / X J 0 , if η[(X J ) out ]>0, indicating that the monitoring result is abnormal, and the feedback information of the eigenvalue of the Jth support vector is issued; otherwise, it means that the intelligent control decision is normal and there is no feedback; Step 2: Based on step 1, the eigenvalues ​​of the remaining M-1 support vectors are input into the deviation coefficient model, and feedback information is issued according to the abnormal monitoring results.

9. An artificial intelligence-based data control method, used to use an artificial intelligence-based data control system according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step 1: Use big data technology to store and update the standard parameters required for data control of target power equipment auxiliary equipment in real time, and build a data control information library for power equipment auxiliary equipment; Step 2: According to the power equipment auxiliary equipment data control information library in step 1, a sensor group is arranged at a key position, and the first target parameter required for the power equipment auxiliary equipment data control is collected in real time according to a set sampling frequency; Step 3: Using big data technology, feature extraction is performed on the first target parameter in step 2 to obtain the second target parameter required for data control of auxiliary equipment of the power equipment; Step 4: Perform feature analysis on the second target parameter in step 3 by using pattern recognition technology and trend prediction technology to obtain feature analysis results; Step 5: Generate corresponding control strategy according to the characteristic analysis results of step 4; Step 6: Monitor the result of the second target parameter after the control strategy of step 5 is executed, and feedback to step 5 based on the abnormal monitoring result.

Images