Intelligent explosion-proof system and method of adaptive environment perception explosion-proof motor

Abstract

The present application relates to the technical field of environment perception and motor control, in particular to an intelligent explosion-proof system and method of an adaptive environment perception explosion-proof motor, the method comprising: defining an occurrence event, taking each dimension of the environment multidimensional data as an event intervention factor, training a multidimensional risk field function, and outputting the risk coefficient of each event intervention factor; generating a risk regulation signal and a standard regulation signal based on the occurrence event and motor safe operation regulation explosion prevention, and constructing a label set of motor parameters, quantifying the regulation signal, obtaining a competition coefficient difference, constructing a signal competition model based on the label set and the competition coefficient difference, and outputting the regulation parameter of the motor through the signal competition model; the present application enhances the adaptive ability of the motor to complex working conditions through dynamic calculation of risk weight and training of a multidimensional risk field function; the present application also solves the problem of strategy rigidity of traditional methods, and realizes the intelligent balance of motor safe operation and efficient regulation.

Description

Technical Field

[0001] This invention relates to the field of environmental sensing and motor control technology, specifically to an intelligent explosion-proof system and method for an adaptive environmental sensing explosion-proof motor. Background Technology

[0002] Explosion-proof motors are special motors designed for use in flammable and explosive environments, such as petrochemical and coal mine environments. Their design uses technologies such as explosion-proof enclosures, sealed structures, or inert gas protection to prevent internal electrical sparks, high temperatures, or mechanical failures from igniting external explosive gases, dust, or vapors.

[0003] The core purpose of controlling explosion-proof motors is to dynamically ensure safety. In flammable and explosive environments, parameters such as gas concentration, temperature, and humidity may change rapidly with production activities or leakage events. Traditional explosion-proof motors are insufficient in terms of dynamic environmental perception and analysis, making it difficult to capture potential risks in complex environments in real time. Secondly, existing explosion-proof control methods lack flexible decision-making mechanisms, making it difficult to balance the conflict between real-time risks and preset safety standards. Therefore, an intelligent explosion-proof method for adaptive environmental perception explosion-proof motors is needed. Summary of the Invention

[0004] The purpose of this invention is to address the problems existing in the background technology by proposing an intelligent explosion-proof system and method for an adaptive environmental sensing explosion-proof motor.

[0005] The technical solution of the present invention is an intelligent explosion-proof method for an adaptive environment-aware explosion-proof motor, comprising the following steps: defining an event period, acquiring multi-dimensional data of the motor and multi-dimensional data of the environment within the event period, and constructing an event based on the multi-dimensional data of the environment within the event period;

[0006] Each dimension of the multidimensional environmental data is used as an event intervention factor to perform event analysis on the events that occur, obtain the analysis results of the events, train a multidimensional risk field function based on the analysis results, and output the risk coefficient of each event intervention factor;

[0007] Based on recent events, each event association set is determined, the event association sets are processed to determine the environmental risk threshold, the average risk value of the event cycle is obtained, and the environmental risk threshold and the average risk value are combined for analysis to generate risk control signals.

[0008] Obtain the motor safety operation regulations, obtain the standard parameters corresponding to each motor parameter according to the safety operation regulations, compare and analyze the motor parameters with the standard parameters, and generate standard control signals;

[0009] A tag set for motor parameters is constructed based on risk control signals and standard control signals. The control signals are quantified to obtain the competition coefficient of each control signal. The difference in competition coefficients between control signals is obtained, and a set of competition conditions is generated based on the tag set and the difference in competition coefficients.

[0010] A signal competition model is constructed based on a set of competition conditions. The motor control parameters are output through the signal competition model, and the motor parameters are controlled in real time based on the control competition model.

[0011] Preferred methods for performing event analysis on occurring events include:

[0012] The event intervention factors are numbered, and the event intervention factors are arranged and combined to construct intervention factor pairs with two time intervention factors as combinations; wherein, the intervention factor pairs include those obtained by combining two identical event intervention factors;

[0013] The data corresponding to each event intervention factor is normalized, and the asynchronously sampled data is interpolated to align with the timestamp. Noise is removed by using a filter method.

[0014] The covariance of each intervention factor pair is calculated using the following formula across multiple event cycles:

[0015] ;

[0016] In the formula, Let be the covariance of the intervention factor pair (j, k); j and k are the numbers of the event intervention factors, both of which are positive integers, j, k∈[1, p], and p is the total number of event intervention factors; and The mean readings of event intervention factors numbered j and k within the event period are calculated based on the mean of each reading of the j-th sensor and the k-th sensor within the event period, respectively. and , j and k are the average values ​​of the event intervention factors over multiple event periods, respectively; N is the number of event periods.

[0017] Preferably, the covariance matrix is ​​constructed based on the covariance of different intervention factor pairs as follows:

[0018] ;

[0019] Perform eigenvalue decomposition on the covariance matrix to obtain p eigenvalues, and then sort the p eigenvalues ​​in descending order to obtain the feature sequence; label the eigenvalues ​​in the feature sequence as... , eigenvalue The corresponding feature vector is labeled as feature vector. Where h is the feature number, h is a positive integer, and h∈[1,p].

[0020] Preferably, the feature values ​​are sorted according to the sequence element order based on the feature sequence. and the corresponding feature vector They form a principal component;

[0021] The eigenvalue is calculated using the following formula. Contribution rate of a single principal component ;

[0022] ;

[0023] For eigenvalues Principal component contribution rate Perform the calculations to determine the minimum number of accumulations.

[0024] The preferred method for determining the minimum number of accumulations is as follows:

[0025] The cumulative contribution rate is obtained by sequentially summing the principal component contribution rates corresponding to the eigenvalues ​​in the feature sequence. The current feature value is counted and marked as the accumulation count m. The cumulative contribution rate is determined at each accumulation. Does it meet the following requirements: ≥ ;in, The cumulative contribution threshold;

[0026] If the cumulative contribution rate of the first m principal components corresponding to the current accumulation count m is... satisfy: ≥ Then mark the current accumulation count as the minimum accumulation count zm;

[0027] If the cumulative contribution rate of the first m principal components corresponding to the current accumulation count m is... Not satisfied: ≥ Then, the accumulation continues and the current accumulation count m is updated until the cumulative contribution rate of the first m principal components corresponding to the current accumulation count m is reached. satisfy: ≥ .

[0028] Preferably, the absolute weight of the i-th event intervention factor is calculated using the following formula. :

[0029] ;

[0030] In the formula, The eigenvector component magnitude of the eigenvector of the i-th event intervention factor in the h-th principal component;

[0031] The absolute weights are normalized to obtain the risk weights of the event intervention factors. The formula is as follows:

[0032] In the formula, Risk weights for the output event intervention factors;

[0033] Based on the risk weight of each event intervention factor This generates the analysis results of the event analysis process;

[0034] The method for training a multidimensional risk field function based on the analysis results is as follows:

[0035] Through formula Determine the risk coefficient for each event intervention factor. ;

[0036] In the formula, i is the sensor number, i is a positive integer, and i∈[1, p];

[0037] The expression for the multidimensional risk field function is as follows:

[0038] ;

[0039] In the formula, R represents the environmental risk value; The expression is .

[0040] Preferably, the method for constructing the label set of motor parameters includes:

[0041] Based on the risk control confirmation signal, perform the following risk-related actions:

[0042] Risk coefficient The corresponding event intervention factors are sorted in descending order to obtain the intervention factor risk sequence. Based on the sorting order of the intervention factor sequence, the motor control instructions for the event intervention factors are obtained in sequence. The corresponding motor parameters are obtained according to the control instructions and marked as the first target parameters. A first control label S1 is added to the first target parameters.

[0043] Cancel risk operations based on risk control cancellation signals;

[0044] Perform standard operation according to the standard control confirmation signal: add the second control label S2 to the second target parameter;

[0045] Standard operations are cancelled based on standard control signals.

[0046] Preferred methods for quantifying control signals include:

[0047] The first competition coefficient C1 of the risk control signal is determined by the following formula:

[0048] In the formula, This represents the average environmental risk value over the event period. Environmental risk threshold;

[0049] The second competition coefficient C2 of the risk control signal is determined using the following formula:

[0050] In the formula, These are the motor parameters; These are the standard parameters corresponding to the motor parameters;

[0051] The competition coefficient difference is calculated using the formula C0=C1-C2;

[0052] A tag set E for motor parameters is constructed based on the risk control signal and the standard control signal. The tag set E and the competition coefficient difference are then arranged in a non-repeating permutation and combination to obtain the competition condition set R.

[0053] The following combinations of competition condition set R exist:

[0054] R1={E=S1, C0>0}, R2={E=S1, C0<0}, R3={E=S2, C0>0}, R4={E=S2, C0<0}, R5={E= (S1, S2), C0>0}, R6={E= (S1, S2), C0<0}.

[0055] The preferred expression for the signal contention model is as follows:

[0056] ;

[0057] In the formula, These are the adjustment parameters for the motor. The maximum value of the first target parameter;

[0058] Update the risk control signal and standard control signal after adjustment.

[0059] This invention also discloses an intelligent explosion-proof system for an adaptive environment-sensing explosion-proof motor, which applies the aforementioned intelligent explosion-proof method for an adaptive environment-sensing explosion-proof motor, specifically including:

[0060] The event definition module is used to define the event period. Within the event period, it acquires multidimensional data of the motor and multidimensional data of the environment, and constructs an event based on the multidimensional data of the environment within the event period.

[0061] The risk field training module is used to treat each dimension of the multidimensional environmental data as an event intervention factor, perform event analysis on the events that occur, obtain the analysis results of the events, train the multidimensional risk field function based on the analysis results, and output the risk coefficient of each event intervention factor.

[0062] The risk control signal generation module is used to determine the association set of each event based on recent events, process the event association set, determine the environmental risk threshold, obtain the average risk value of the event cycle, and combine the environmental risk threshold and the average risk value for analysis to generate a risk control signal.

[0063] The standard control signal generation module is used to obtain the motor safety operation regulations, obtain the standard parameters corresponding to each motor parameter according to the safety operation regulations, compare and analyze the motor parameters with the standard parameters, and generate standard control signals.

[0064] The competition analysis module is used to construct a tag set of motor parameters based on risk control signals and standard control signals, quantify the control signals, obtain the competition coefficient of each control signal, acquire the difference in competition coefficients between control signals, and generate a competition condition set based on the tag set and the difference in competition coefficients.

[0065] The parameter control module is used to construct a signal competition model based on a set of competition conditions, output the motor control parameters through the signal competition model, and perform real-time control of the motor parameters based on the control competition model.

[0066] Compared with the prior art, the above-mentioned technical solution of the present invention has the following beneficial technical effects:

[0067] (1) By defining the event cycle, integrating multi-dimensional data of motor and environment, and constructing covariance matrix for principal component analysis, key risk factors are effectively extracted. By dynamically calculating risk weights and training multi-dimensional risk field functions, the assessment accuracy of environmental risk values ​​is significantly improved, and the adaptability of explosion-proof motor to complex working conditions is enhanced.

[0068] (2) By constructing a tag set and a competition condition set, and combining the competition coefficient difference between the risk control signal and the standard control signal, a signal competition model was designed. This model can quantify the control priority in real time and dynamically output the optimal control parameters, solving the problems of control lag and rigid strategy in traditional methods, and realizing the intelligent balance between safe operation and efficient control of motor. Attached Figure Description

[0069] Figure 1 This is a block diagram of a method according to Embodiment 1 of the present invention. Detailed Implementation

[0070] Example 1, as Figure 1As shown, the present invention proposes an intelligent explosion-proof method for an adaptive environment-sensing explosion-proof motor, comprising the following steps:

[0071] Define an event period. Within the event period, acquire multidimensional data of the motor and the environment, and construct an event based on the environmental multidimensional data within the event period.

[0072] Each dimension of the multidimensional environmental data is used as an event intervention factor to perform event analysis on the events that occur, obtain the analysis results of the events, train a multidimensional risk field function based on the analysis results, and output the risk coefficient of each event intervention factor;

[0073] Methods for conducting event analysis on events include:

[0074] The event intervention factors are numbered, and the event intervention factors are arranged and combined to construct intervention factor pairs with two time intervention factors as combinations; wherein, the intervention factor pairs include those obtained by combining two identical event intervention factors;

[0075] The data corresponding to each event intervention factor is normalized, and the asynchronously sampled data is interpolated to align with the timestamp. Noise is removed by using a filter method.

[0076] The covariance of each intervention factor pair is calculated using the following formula across multiple event cycles:

[0077] ;

[0078] In the formula, Let be the covariance of the intervention factor pair (j, k); j and k are the numbers of the event intervention factors, both of which are positive integers, j, k∈[1, p], and p is the total number of event intervention factors; and The mean readings of event intervention factors numbered j and k within the event period are calculated based on the mean of each reading of the j-th sensor and the k-th sensor within the event period, respectively. and These are the average values ​​of the event intervention factors numbered j and k over multiple event periods, respectively; N is the number of event periods.

[0079] The covariance matrix is ​​constructed based on the covariance of different intervention factor pairs as follows:

[0080] ;

[0081] Perform eigenvalue decomposition on the covariance matrix to obtain p eigenvalues, and then sort the p eigenvalues ​​in descending order to obtain the feature sequence; label the eigenvalues ​​in the feature sequence as... , eigenvalue The corresponding feature vector is labeled as feature vector. Where h is the feature number, h is a positive integer, and h∈[1, p];

[0082] Based on the feature sequence, the feature values ​​are sorted according to the sequence element order. and the corresponding feature vector They form a principal component;

[0083] The eigenvalue is calculated using the following formula. Contribution rate of a single principal component ;

[0084] ;

[0085] For eigenvalues Principal component contribution rate Perform the calculations to determine the minimum number of accumulations;

[0086] The method for determining the minimum number of accumulations is as follows:

[0087] The cumulative contribution rate is obtained by sequentially summing the principal component contribution rates corresponding to the eigenvalues ​​in the feature sequence. The current feature value is counted and marked as the accumulation count m. The cumulative contribution rate is determined at each accumulation. Does it meet the following requirements: ≥ ;in, The cumulative contribution threshold;

[0088] If the cumulative contribution rate of the first m principal components corresponding to the current accumulation count m is... satisfy: ≥ Then mark the current accumulation count as the minimum accumulation count zm;

[0089] If the cumulative contribution rate of the first m principal components corresponding to the current accumulation count m is... Not satisfied: ≥ Then, the accumulation continues and the current accumulation count m is updated until the cumulative contribution rate of the first m principal components corresponding to the current accumulation count m is reached. satisfy: ≥ ;

[0090] The absolute weight of the i-th event intervention factor is calculated using the following formula. :

[0091] ;

[0092] In the formula, The eigenvector component magnitude of the eigenvector of the i-th event intervention factor in the h-th principal component;

[0093] The absolute weights are normalized to obtain the risk weights of the event intervention factors. The formula is as follows:

[0094] In the formula, Risk weights for the output event intervention factors;

[0095] Based on the risk weight of each event intervention factor This generates the analysis results of the event analysis process;

[0096] The method for training a multidimensional risk field function based on the analysis results is as follows:

[0097] Through formula Determine the risk coefficient for each event intervention factor. ;

[0098] In the formula, i is the sensor number, i is a positive integer, and i∈[1, p];

[0099] The expression for the multidimensional risk field function is as follows:

[0100] ;

[0101] In the formula, R represents the environmental risk value; The expression is ;

[0102] Based on recent events, each event association set is determined, the event association sets are processed to determine the environmental risk threshold, the average risk value of the event cycle is obtained, and the environmental risk threshold and the average risk value are combined for analysis to generate risk control signals.

[0103] After each event ends, the event is marked as the target event. The first x events of the target event are combined with the target event to obtain the event association set corresponding to the target event.

[0104] Based on whether the motor explodes within the event cycle, the event result of the event is defined, and regular events and abnormal events are obtained. Event association sets of multiple abnormal events are obtained and all are marked as abnormal event association sets.

[0105] Obtain the environmental risk value at different times within the event period corresponding to each event in the abnormal event association set to obtain the abnormal risk value set of the abnormal event association set. Calculate the mean of the abnormal risk value sets of multiple abnormal event association sets, and use the calculation result as the environmental risk threshold. ;

[0106] The average environmental risk value of the event is obtained by averaging the environmental risk values ​​at different times within the event period. ;

[0107] Average environmental risk value Environmental risk threshold Data comparison was performed, and the average environmental risk value was... Greater than the environmental risk threshold Then a risk control confirmation signal is generated; if the average environmental risk value Not greater than the environmental risk threshold If so, no action will be taken and a risk control cancellation signal will be generated;

[0108] Average environmental risk value Perform real-time updates;

[0109] Obtain the motor safety operation regulations, obtain the standard parameters corresponding to each motor parameter according to the safety operation regulations, compare and analyze the motor parameters with the standard parameters, and generate standard control signals;

[0110] The methods for analyzing data for each motor parameter according to safe operation regulations include:

[0111] Standard operating parameters corresponding to motor parameters Perform data comparison and generate comparison results for each motor parameter;

[0112] If motor parameters are greater than standard operating parameters If the motor parameter is not greater than the standard operating parameter, then mark it as the second target parameter and generate a standard control confirmation signal; If so, no operation will be performed and a standard control cancellation signal will be generated;

[0113] A tag set for motor parameters is constructed based on risk control signals and standard control signals. The control signals are quantified to obtain the competition coefficient of each control signal. The difference in competition coefficients between control signals is obtained, and a set of competition conditions is generated based on the tag set and the difference in competition coefficients.

[0114] Methods for constructing a label set for motor parameters include:

[0115] Based on the risk control confirmation signal, perform the following risk-related actions:

[0116] Risk coefficient The corresponding event intervention factors are sorted in descending order to obtain the intervention factor risk sequence. Based on the sorting order of the intervention factor sequence, the motor control instructions for the event intervention factors are obtained in sequence. The corresponding motor parameters are obtained according to the control instructions and marked as the first target parameters. A first control label S1 is added to the first target parameters.

[0117] Cancel risk operations based on risk control cancellation signals;

[0118] Perform standard operation according to the standard control confirmation signal: add the second control label S2 to the second target parameter;

[0119] Cancel standard operations based on standard control signals;

[0120] Methods for quantifying control signals include:

[0121] The first competition coefficient C1 of the risk control signal is determined by the following formula:

[0122] ;

[0123] In the formula, This represents the average environmental risk value over the event period. Environmental risk threshold;

[0124] The second competition coefficient C2 of the risk control signal is determined using the following formula:

[0125] ;

[0126] In the formula, These are the motor parameters; These are the standard parameters corresponding to the motor parameters;

[0127] The competition coefficient difference is calculated using the formula C0=C1-C2;

[0128] A tag set E for motor parameters is constructed based on the risk control signal and the standard control signal. The tag set E and the competition coefficient difference are then arranged in a non-repeating permutation and combination to obtain the competition condition set R.

[0129] The possible combinations of the competition condition set R are as follows:

[0130] R1={E=S1, C0>0}, R1={E=S1, C0<0}, R3={E=S2, C0>0}, R4={E=S2, C0<0}, R5={E= (S1, S2), C0>0}, R6={E= (S1, S2), C0<0};

[0131] A signal competition model is constructed based on the competition condition set. The motor control parameters are output through the signal competition model, and the motor parameters are controlled in real time based on the control competition model.

[0132] The expression for the signal contention model is as follows:

[0133] ;

[0134] In the formula, These are the adjustment parameters for the motor. The maximum value of the first target parameter;

[0135] Update the risk control signal and standard control signal after adjustment.

[0136] Example 2: The intelligent explosion-proof system for an adaptive environment-sensing explosion-proof motor proposed in this invention is applied to the intelligent explosion-proof method for an adaptive environment-sensing explosion-proof motor proposed in Example 1, and specifically includes:

[0137] The event definition module is used to define the event period. Within the event period, it acquires multidimensional data of the motor and multidimensional data of the environment, and constructs an event based on the multidimensional data of the environment within the event period.

[0138] The risk field training module is used to treat each dimension of the multidimensional environmental data as an event intervention factor, perform event analysis on the events that occur, obtain the analysis results of the events, train the multidimensional risk field function based on the analysis results, and output the risk coefficient of each event intervention factor.

[0139] The risk control signal generation module is used to determine the association set of each event based on recent events, process the event association set, determine the environmental risk threshold, obtain the average risk value of the event cycle, and combine the environmental risk threshold and the average risk value for analysis to generate a risk control signal.

[0140] The standard control signal generation module is used to obtain the motor safety operation regulations, obtain the standard parameters corresponding to each motor parameter according to the safety operation regulations, compare and analyze the motor parameters with the standard parameters, and generate standard control signals.

[0141] The competition analysis module is used to construct a tag set of motor parameters based on risk control signals and standard control signals, quantify the control signals, obtain the competition coefficient of each control signal, obtain the difference in competition coefficients between control signals, and generate a set of competition conditions based on the tag set and the difference in competition coefficients.

[0142] The parameter control module is used to construct a signal competition model based on a set of competition conditions, output the motor control parameters through the signal competition model, and perform real-time control of the motor parameters based on the control competition model.

[0143] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A smart explosion-proof method of an adaptive environment-aware explosion-proof motor, characterized in that, The method comprises the following steps: Defining an event period, obtaining motor multi-dimensional data and environment multi-dimensional data in the event period, and constructing an occurrence event based on the environment multi-dimensional data in the event period; Each dimension of the environment multi-dimensional data is taken as an event intervention factor, and event analysis is performed on the occurrence event to obtain an analysis result of the occurrence event, a multi-dimensional risk field function is trained based on the analysis result, and a risk coefficient of each event intervention factor is output; Determination of absolute weight of each event intervention factor based on principal component analysis method : ； In the formula, The eigenvector component module length of the eigenvector of the ith event intervention factor in the hth principal component; m represents the minimum cumulative number determined by the cumulative contribution rate threshold value; The single principal component contribution rate of the eigenvalue ; and The absolute weight of the ith event intervention factor; The absolute weight is normalized to obtain a risk weight of the event intervention factor, and the processing formula is as follows: ; where, is the risk weight of the outputted event intervention factor; According to the risk weight of each event intervention factor generating an analysis result of the event analysis process; The method for training the multi-dimensional risk field function based on the analysis result is as follows: The risk coefficient of each event intervention factor is determined by the formula ;​ In the formula, i is the sensor number, i is a positive integer, i∈[1, p]; p is the total number of event intervention factors; S i the average of the readings of the i-th event intervention factor over the event period; The expression of the multi-dimensional risk field function is as follows: ； wherein R is an environmental risk value; The expression is ; Based on the recent occurrence event, each event association set is determined, the event association set is processed, the environmental risk threshold is determined, the average risk value of the event period is obtained, the environmental risk threshold and the average risk value are combined for analysis, and a risk control signal is generated; The motor safety operation regulation is obtained, the standard parameters corresponding to each motor parameter are obtained according to the safety operation regulation, the motor parameters are compared and analyzed with the standard parameters, and a standard control signal is generated; A label set of the motor parameters is constructed according to the risk control signal and the standard control signal, the control signal is quantized, the competition coefficient of each control signal is obtained, the competition coefficient difference between the control signals is obtained, and a competition condition set is generated based on the label set and the competition coefficient difference; A signal competition model is constructed based on the competition condition set, the control parameters of the motor are output through the signal competition model, and the motor parameters are controlled in real time based on the control competition model.

2. The intelligent explosion-proof method of the self-adaptive environment-aware explosion-proof motor according to claim 1, characterized in that, The method for performing event analysis on the occurrence event comprises: The event intervention factors are numbered, the event intervention factors are arranged and combined, and an intervention factor pair combined with two time intervention factors is constructed; wherein the intervention factor pair is obtained by combining two same event intervention factors; The data corresponding to each event intervention factor is normalized, and the asynchronously sampled data is interpolated to align the time stamp, and the noise is removed by using a filter method; In a plurality of event periods, the covariance of each intervention factor pair is calculated by the following formula: ； wherein, Cov(j, k) is the covariance of the intervention factors pair (j, k); j and k are the numbers of the event intervention factors, j and k are positive integers, j, k ∈ [1, p], p is the total number of event intervention factors; and and and are the means of the event intervention factors numbered j and k over the event period, respectively based on the mean calculation of each reading of the jth type of sensor and the kth type of sensor over the event period; and 3. The intelligent explosion-proof method of the self-adaptive environment-aware explosion-proof motor according to claim 2, characterized in that, A covariance matrix is constructed according to the covariances of different intervention factor pairs as follows: ； The covariance matrix is decomposed to obtain p characteristic values, and the p characteristic values are arranged in descending order to obtain a characteristic sequence; Mark the feature value in the feature sequence as Mark the feature value The corresponding feature vector is marked as feature vector ; wherein h is a feature number, h is a positive integer, h∈[1, p].

4. The intelligent explosion-proof method of the self-adaptive environment-aware explosion-proof motor according to claim 3, characterized in that, the feature values based on the sequence of features in the sequence element ordering and the corresponding feature vectors consisting of one principal component; The eigenvalue is calculated by the following equation The individual principal component contribution rate of the feature value ; ； The principal component contribution rate of the characteristic value The principal component contribution rate of the characteristic value The minimum cumulative number of times is determined by performing an operation.

5. The intelligent explosion-proof method of the self-adaptive environment-aware explosion-proof motor according to claim 4, characterized in that, The method for determining the minimum cumulative number comprises: The principal component contribution rate corresponding to the feature value in the feature sequence is sequentially accumulated to obtain a cumulative contribution rate The number of the current feature value is counted and marked as the accumulation number m, and the cumulative contribution rate is judged at each accumulation Whether the following condition is met: ≥ ; wherein, is the cumulative contribution threshold value; If the cumulative contribution rate of the first m principal components corresponding to the current cumulative number m is less than the preset threshold value, the current cumulative number m is marked as the minimum cumulative number zm. Satisfying: ≥ The current cumulative number is marked as the minimum cumulative number zm. If the cumulative contribution rate of the first m principal component contribution rates corresponding to the current cumulative number m Does not satisfy: ≥ Then the accumulation is continuously carried out and the current cumulative number m is updated until the cumulative contribution rate of the first m principal component contribution rates corresponding to the current cumulative number m Satisfies: ≥ .

6. The intelligent explosion-proof method of the self-adaptive environment-aware explosion-proof motor according to claim 5, characterized in that, The method for constructing the label set of the motor parameters comprises: The following risk operation is performed according to the risk control confirmation signal: Risk coefficient The corresponding event intervention factors are sorted in descending order to obtain an intervention factor risk sequence, the motor control instructions for the event intervention factors are obtained in sequence based on the intervention factor sequence sorting order, and the corresponding motor parameters are obtained according to the control instructions and marked as first target parameters; a first control label S1 is added to the first target parameters; The risk operation is cancelled according to the risk control cancellation signal; The standard operation is performed according to the standard control confirmation signal: a second control label S2 is added to the second target parameter; The standard operation is cancelled according to the standard control signal.

7. The intelligent explosion-proof method of the self-adaptive environment-aware explosion-proof motor according to claim 6, characterized in that, The method for quantizing the control signal comprises: The first competition coefficient C1 of the risk control signal is determined by the following formula: ; wherein is an average environmental risk value for the event period; is an environmental risk threshold value; The second competition coefficient C2 of the risk control signal is determined by the following formula: ; wherein are motor parameters; are standard parameters corresponding to the motor parameters; The competition coefficient difference is calculated by the formula C0=C1-C2; According to the risk control signal and the standard control signal, a label set E of motor parameters is constructed, and the label set E and the difference in competition coefficients are non-repeatedly arranged and combined to obtain a competition condition set R; The combinations of the competition condition set R are as follows: R1={E=S1, C0>0}, R2={E=S1, C0<0}, R3={E=S2, C0>0}, R4={E=S2, C0<0}, R5={E=(S1, S2), C0>0}, and R6={E=(S1, S2), C0<0}.

8. The intelligent explosion-proof method of the self-adaptive environment-aware explosion-proof motor according to claim 7, characterized in that, The expression of the signal competition model is as follows: ； In the formula, is a control parameter for the motor parameter, is a first target parameter rated maximum value; The risk control signal and the standard control signal after regulation are updated.

9. An intelligent explosion-proof system of an adaptive environment-sensing explosion-proof motor, applied to the intelligent explosion-proof method of any one of claims 1 to 8, characterized in that, Specifically, the method comprises the following steps: An event definition module is used to define an event period, and in the event period, multi-dimensional motor data and multi-dimensional environmental data are obtained, and an occurrence event is constructed based on the multi-dimensional environmental data in the event period; A risk field training module is used to take each dimension of the multi-dimensional environmental data as an event intervention factor, perform event analysis on the occurrence event, obtain an analysis result of the occurrence event, train a multi-dimensional risk field function based on the analysis result, and output a risk coefficient of each event intervention factor; A risk control signal generation module is used to determine each event association set based on a recent occurrence event, process the event association set, determine an environmental risk threshold, obtain an average risk value of the event period, analyze the environmental risk threshold and the average risk value, and generate a risk control signal; A standard control signal generation module is used to obtain motor safe operation regulations, obtain standard parameters corresponding to each motor parameter according to the safe operation regulations, compare and analyze the motor parameters and the standard parameters, and generate a standard control signal; A competition analysis module is used to construct a label set of motor parameters according to the risk control signal and the standard control signal, quantify the control signals, obtain competition coefficients of each control signal, obtain the difference in competition coefficients between the control signals, and generate a competition condition set based on the label set and the difference in competition coefficients A parameter control module is used to construct a signal competition model based on the competition condition set, output motor control parameters through the signal competition model, and perform real-time control on the motor parameters based on the control competition model.

Images