Disclosure of Invention
(One) solving the technical problems
Aiming at the defects of the prior art, the invention provides an intelligent monitoring system for the running state of a switch cabinet based on IEC61850 communication, which has the advantages of real time, accuracy and the like, and solves the problems of excessively relying on manual information and lacking self-learning capability.
(II) technical scheme
In order to solve the technical problem that the artificial information is excessively depended and the self-learning capability is lacking, the invention provides the following technical scheme:
the invention discloses an intelligent monitoring method for the running state of a switch cabinet based on IEC61850 communication, which specifically comprises the following steps:
S1, installing a series of state parameter acquisition devices on an internal circuit node and an external circuit node of a switch cabinet, and acquiring state parameter data in the switch cabinet in real time based on the installed state parameter acquisition devices and storing the state parameter data in a database; the state parameter acquisition device includes: a current state monitor, a voltage state monitor, and a switchgear temperature state monitor;
S2, extracting historical state parameter data in a switch cabinet stored in a database, and initializing the extracted historical state parameter data to obtain initialized historical state parameter data; the initialized historical state parameter data comprise switch cabinet state parameter data collected by state parameter collection equipment and control instruction data in corresponding time; the control instruction data in the corresponding time comprises a current control instruction and a voltage control instruction;
S3, constructing a relation model of a control instruction and the state parameters of the switch cabinet based on the state parameters of the switch cabinet collected by the state parameter collection equipment in the initialized historical state parameter data and the control instruction data in the corresponding time;
s4, analyzing and judging a relation model of the control command and the state parameter of the switch cabinet based on the BP neural network, and determining the relation model of the control command and the state parameter of the switch cabinet;
S5, after analysis is completed, predicting state parameter data in the switch cabinet after the control instruction is executed through a determined relation model of the control instruction and the state parameters in the switch cabinet, and uploading the predicted state parameter data in the switch cabinet after the control instruction is executed through an IEC61850 communication mode;
S6, intelligent monitoring is carried out on the running state of the switch cabinet based on state parameter data in the switch cabinet after the predicted control instruction uploaded in the IEC61850 communication mode is executed.
According to the invention, a series of state parameter acquisition devices are installed on an internal circuit node of the switch cabinet, and state parameter data in the switch cabinet are acquired in real time based on the installed state parameter acquisition devices; meanwhile, historical state parameter data in the switch cabinet are collected, the collected historical state parameter data are initialized, a relation model of a control instruction and the switch cabinet state parameter is built based on the switch cabinet state parameter collected by state parameter collecting equipment in the initialized historical state parameter data and control instruction data in corresponding time, meanwhile, the relation model is analyzed and judged through a BP neural network, the state parameter of the switch cabinet is predicted based on a control instruction generated in real time through the analyzed relation model, intelligent monitoring of the operation state of the switch cabinet is completed, and accuracy and instantaneity of intelligent monitoring of the operation state of the switch cabinet are guaranteed.
Preferably, the step of extracting the historical state parameter data in the switch cabinet stored in the database and initializing the extracted historical state parameter data includes the steps of:
S21, filtering the collected historical state parameter data to obtain filtered historical state parameter data;
S22, classifying the filtered historical state parameter data to obtain classified historical state parameter data, wherein the classified historical state parameter data comprises standard historical state parameter data and nonstandard historical state parameter data;
s23, setting the filtered and classified historical state parameter data as initialized historical state parameter data.
Preferably, the filtering the collected historical state parameter data to obtain filtered historical state parameter data includes the following steps: according to the circuit principle, in a closed circuit, the circuit states include: a via, a circuit breaker, and a short circuit; setting state parameter data acquired in real time in a channel state as historical state parameter data when a circuit is normal; setting state parameter data acquired in real time in a circuit breaking state and a short circuit state as historical state parameter data when a circuit is abnormal; in the closed circuit, when current flows out from the positive electrode of the power supply, the current passes through the electric appliance and returns to the negative electrode to form a complete loop, and the complete loop is expressed as a passage; in a closed circuit, when there is a completely open line in the circuit, so that current cannot flow through this line, but there are still other paths in the circuit through which current can flow, denoted as open circuit; in a closed circuit, when abnormal connection exists in the circuit, a current larger than rated current flows in the circuit, so that a circuit line is burnt out, and the circuit is expressed as a short circuit; and removing short circuit data which are generated by burning out the circuit in the circuit because the current which is larger than the rated current flows in the circuit and exists in the collected historical state parameter data when the circuit is abnormal, so as to obtain the filtered historical state parameter data.
Preferably, the classifying the filtered historical state parameter data includes the steps of: setting a standard state range threshold value in the running process of the switch cabinet, and classifying state parameter data acquired in real time under a passage state based on the set standard state range threshold value; the method comprises the steps of setting the passage state data in the closed circuit within the standard state range threshold as standard historical state parameter data, and setting the passage state data in the closed circuit not within the standard state range threshold as non-standard historical state parameter data.
According to the invention, the collected historical state parameter data in the switch cabinet is initialized, and the validity of the historical state parameter data is improved and the accuracy of intelligent monitoring of the operation state of the switch cabinet is ensured through filtering and classifying operations.
Preferably, the construction of the relation model of the control instruction and the switch cabinet state parameter based on the switch cabinet state parameter data collected by the state parameter collecting device in the initialized historical state parameter data and the control instruction data in the corresponding time includes the following steps:
S31, generating a current control instruction and a voltage control instruction based on control instruction data in corresponding time in initialized historical state parameter data;
S32, monitoring the temperature of the switch cabinet under the set current intensity and voltage intensity through a switch cabinet temperature state monitor based on the generated current control instruction and voltage control instruction; recording the temperature change state of the switch cabinet once when the current and the voltage change once;
S33, after repeatedly collecting temperature change data of the switch cabinet when the current and the voltage change, carrying out standardization processing on the collected current, voltage and temperature change data, and establishing a relation model of a control instruction and the state parameters of the switch cabinet.
Preferably, the standardized processing of the collected current, voltage and temperature change data and the establishment of the relation model of the control command and the state parameters of the switch cabinet comprise the following steps: the formula for carrying out standardization processing on the collected current, voltage and temperature change data is as follows: Wherein Z represents data before normalization processing,Representing the normalized data; maximum value in state data representing each parameter variable; A minimum value in the state data representing each parameter variable; the calculation formula of the relation model of the control instruction and the state parameter of the switch cabinet is as follows: the relation model between the temperature state parameter in the switch cabinet and the voltage and current set by the control instruction is as follows: Wherein,In order to control the temperature state parameters in the switch cabinet obtained after the instruction execution,The voltage change amount set by the control command is indicated,The current variation set by the control command is indicated,The amount of change in time is indicated,In order to control the temperature state parameters in the switch cabinet collected before the instruction execution,And the relation weight of the voltage and the current is set for the temperature state parameter and the control instruction in the switch cabinet.
According to the intelligent monitoring method and the intelligent monitoring system, the relation model of the control instruction and the switch cabinet state parameter is built in a standardized mode for the switch cabinet state parameter collected by the state parameter collecting device in the initialized historical state parameter and the control instruction data in the corresponding time, so that the intelligent monitoring stability of the switch cabinet running state is improved.
Preferably, the analysis and judgment are performed on the relation model of the control command and the state parameter of the switch cabinet based on the BP neural network, and the determination of the relation model of the control command and the state parameter of the switch cabinet comprises the following steps:
S41, selecting part of initialized historical state parameter data from all initialized historical state parameter data to form a historical state parameter data training set;
S42, determining the structure of the BP neural network, a learning algorithm and selecting related parameters;
S43, comparing the determined BP neural network with a relation model of the control command and the state parameter of the switch cabinet, analyzing and judging whether the relation model of the control command and the state parameter of the switch cabinet meets the error requirement; when the comparison results in that the relation model of the control command and the state parameter of the switch cabinet meets the error requirement, setting the relation model of the current control command and the state parameter of the switch cabinet as a prediction model of the relation between the control command and the state parameter of the switch cabinet, and predicting the state parameter of the switch cabinet after the control command for acquiring the state parameter data in the switch cabinet in real time is executed by the prediction model to obtain the temperature state parameter in the switch cabinet after the control command is executed; and when the comparison results in that the relation model of the control command and the state parameter of the switch cabinet does not accord with the error requirement, reestablishing the relation model of the control command and the state parameter of the switch cabinet.
Preferably, the comparing the determined BP neural network with the relation model of the control command and the state parameter of the switch cabinet, analyzing and judging whether the relation model of the control command and the state parameter of the switch cabinet meets the error requirement comprises the following steps: the BP neural network structure comprises: an input layer, an hidden layer and an output layer; the learning algorithm of the BP neural network comprises the following steps: setting the input of the input layer asWhereinRepresenting the control instructions and the switch cabinet state parameters in the first historical state parameter data input,The control instruction and the switch cabinet state parameter in the input mth historical state parameter data are represented, and the output of the output layer isWhereinA first predictive switch cabinet temperature state parameter indicative of the output,Representing the temperature state parameter in the output nth prediction switch cabinet, wherein the hidden layer comprises q neurons, v is the weight from the input layer to the hidden layer, and w is the weight from the hidden layer to the output layer; the process of the BP neural network comprises two stages of forward propagation of signals and backward propagation of errors; the forward propagation of the signal is that data is input from an input layer, passes through an hidden layer and finally reaches an output layer; the back propagation of errors is that the data from the output layer to the hidden layer, and finally to the input layer, the weight from the hidden layer to the output layer and the weight from the input layer to the hidden layer are sequentially adjusted; the forward propagation of the signal is calculated as:
the calculation formula for transmitting the historical state parameter data from the input layer to the hidden layer is as follows: Wherein,Representing the input of the h hidden layer neuron,Representing the weight of the ith input of the input layer to the h hidden layer neuron, and m represents the quantity of input data of the input layer; Representing an input layer ith input; Representing the bias value of the h hidden layer neuron; the calculation formula for transmitting the historical state parameter data from the hidden layer to the output layer is as follows: Wherein,Representing the input of the jth output layer neuron,Representing weights of h hidden layer neurons to j output layer neurons, q representing the number of hidden layer neurons; representing the output of the h hidden layer neuron; representing the bias value of the jth output layer neuron; the output layer directly outputs the received historical state parameter data transmitted by the hidden layer after receiving the historical state parameter data transmitted by the hidden layer; i.e.，Representing the output of the jth output layer neuron; calculating an error between the output layer and the expected value, setting an error threshold between the output layer and the expected value, and when the error between the output layer and the expected value is greater than or equal to the set error threshold, sequentially adjusting the weight from the hidden layer to the output layer and the weight from the input layer to the hidden layer through a counter propagation stage of the error; the error calculation formula is as follows: Wherein E is an error, and the error is defined as,N is the number of output neurons, which is the expected value of the j-th output; the calculation formula of the weight adjustment is as follows: Wherein,Representing a weight adjustment value, i representing a learning rate, E representing an error, y representing an output of the output layer; adjusting the weight by means of updating the weight adjustment value of the current weight, and performing iterative computation on the adjusted weight serving as a new weight in forward propagation computation of the signal until a BP neural network model meeting the error requirement is obtained; the BP neural network model meeting the error requirement and the relation model between the control instruction and the state parameter of the switch cabinet are obtained by respectively inputting the control instruction and the state parameter of the switch cabinet, which are collected in real time, and the output result (the temperature state parameter in the switch cabinet is predicted) of the BP neural network model meeting the error requirement and the output result (the temperature state parameter in the switch cabinet is obtained after execution) of the relation model between the control instruction and the state parameter of the switch cabinet are compared, an error threshold is set, and whether the output result of the relation model between the control instruction and the state parameter of the switch cabinet meets the error is judged based on the set error threshold.
According to the invention, the BP neural network is used for analyzing and judging the relation model of the control command and the state parameter of the switch cabinet, so that the accuracy of the relation model of the control command and the state parameter of the switch cabinet is ensured, the accuracy of intelligent monitoring of the running state of the switch cabinet is improved, the BP neural network model meeting the error requirement can obtain relatively long output calculation time, the workload is reduced by searching the relation model similar to the BP neural network model error meeting the error requirement, and the judging efficiency is improved.
The invention also discloses an intelligent monitoring system for the running state of the switch cabinet based on IEC61850 communication, which comprises a data acquisition module, a database, a data processing module, a model construction module, a neural network analysis module, a communication transmission module and a monitoring prediction module; the data acquisition module is used for acquiring the state parameters in the switch cabinet in real time through installed state parameter acquisition equipment; the state parameter acquisition device includes: a current state monitor, a voltage state monitor, and a switchgear temperature state monitor; the current state monitor is used for monitoring the current state of the switch cabinet in real time; the voltage state monitor is used for monitoring the voltage state of the switch cabinet in real time; the switch cabinet temperature state monitor is used for monitoring the temperature state of the switch cabinet in real time; the data processing module is used for filtering and classifying the acquired data; the database is used for storing state parameter data acquired in real time; the model construction module is used for establishing a relation model between the control instruction and the state parameters of the switch cabinet; the neural network analysis module is used for training state parameters in the switch cabinet; the monitoring prediction module is used for monitoring and predicting state parameters in the switch cabinet; the communication transmission module is used for transmitting the monitoring prediction data of the state parameters in the switch cabinet.
(III) beneficial effects
Compared with the prior art, the invention provides the intelligent monitoring system and the intelligent monitoring method for the operation state of the switch cabinet based on IEC61850 communication, which have the following beneficial effects:
1. According to the invention, a series of state parameter acquisition devices are installed on an internal circuit node of the switch cabinet, and state parameter data in the switch cabinet are acquired in real time based on the installed state parameter acquisition devices; meanwhile, historical state parameter data in the switch cabinet are collected, the collected historical state parameter data are initialized, a relation model of a control instruction and the switch cabinet state parameter is built based on the switch cabinet state parameter collected by state parameter collecting equipment in the initialized historical state parameter data and control instruction data in corresponding time, meanwhile, analysis and judgment are carried out on the relation model through a BP neural network, the state parameter in the switch cabinet is predicted based on a control instruction generated in real time through the analyzed relation model, intelligent monitoring on the operation state of the switch cabinet is completed, and accuracy and instantaneity of intelligent monitoring on the operation state of the switch cabinet are guaranteed;
2. according to the method, the collected historical state parameter data in the switch cabinet are initialized, and the validity of the historical state parameter data is improved through filtering and classifying operation, so that the accuracy of intelligent monitoring of the operation state of the switch cabinet is ensured;
3. According to the intelligent monitoring method, the relation model of the control instruction and the switch cabinet state parameter is constructed in a standardized mode for the switch cabinet state parameter collected by the state parameter collecting equipment in the initialized historical state parameter and the control instruction data in the corresponding time, so that the intelligent monitoring stability of the switch cabinet running state is improved;
4. According to the invention, the BP neural network is used for analyzing and judging the relation model of the control command and the state parameter of the switch cabinet, so that the accuracy of the relation model of the control command and the state parameter of the switch cabinet is ensured, the accuracy of intelligent monitoring of the running state of the switch cabinet is improved, the BP neural network model meeting the error requirement can obtain relatively long output calculation time, the workload is reduced by searching the relation model similar to the BP neural network model error meeting the error requirement, and the judging efficiency is improved.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention discloses an intelligent monitoring method for the running state of a switch cabinet based on IEC61850 communication, which specifically comprises the following steps:
S1, installing a series of state parameter acquisition devices on an internal circuit node and an external circuit node of a switch cabinet, and acquiring state parameter data in the switch cabinet in real time based on the installed state parameter acquisition devices; the state parameter acquisition device includes: a current state monitor, a voltage state monitor, and a switchgear temperature state monitor;
S2, extracting historical state parameter data in a switch cabinet stored in a database, and initializing the extracted historical state parameter data to obtain initialized historical state parameter data; the initialized historical state parameter data comprise switch cabinet state parameter data collected by state parameter collection equipment and control instruction data in corresponding time;
S21, filtering the collected historical state parameter data to obtain filtered historical state parameter data; according to the circuit principle, in a closed circuit, the circuit states include: a via, a circuit breaker, and a short circuit; setting state parameter data acquired in real time in a channel state as historical state parameter data when a circuit is normal; setting state parameter data acquired in real time in a circuit breaking state and a short circuit state as historical state parameter data when a circuit is abnormal; in the closed circuit, when current flows out from the positive electrode of the power supply, the current passes through the electric appliance and returns to the negative electrode to form a complete loop, and the complete loop is expressed as a passage; in a closed circuit, when there is a completely open line in the circuit, so that current cannot flow through this line, but there are still other paths in the circuit through which current can flow, denoted as open circuit; in a closed circuit, when abnormal connection exists in the circuit, a current larger than rated current flows in the circuit, so that a circuit line is burnt out, and the circuit is expressed as a short circuit; further, the filtered historical state parameter data is obtained by removing short circuit data which are generated by burning out the circuit in the circuit because the current which is larger than the rated current flows in the circuit and exists in the collected historical state parameter data when the circuit is abnormal;
S22, classifying the filtered historical state parameter data to obtain classified historical state parameter data, wherein the classified historical state parameter data comprises standard historical state parameter data and nonstandard historical state parameter data; setting a standard state range threshold value in the running process of the switch cabinet, and classifying state parameter data acquired in real time under a passage state based on the set standard state range threshold value; further, setting the passage state data in the closed circuit within the standard state range threshold as standard historical state parameter data, and setting the passage state data in the closed circuit not within the standard state range threshold as non-standard historical state parameter data;
S23, setting the filtered and classified historical state parameter data as initialized historical state parameter data;
S3, constructing a relation model of a control instruction and the state parameters of the switch cabinet based on the state parameter data of the switch cabinet collected by the state parameter collecting equipment in the initialized historical state parameter data and the control instruction data in the corresponding time; the control instruction data in the corresponding time comprises a current control instruction and a voltage control instruction;
S31, generating a current control instruction and a voltage control instruction based on control instruction data in corresponding time in initialized historical state parameter data;
S32, monitoring the temperature of the switch cabinet under the set current intensity and voltage intensity through a switch cabinet temperature state monitor based on the generated current control instruction and voltage control instruction; recording the temperature change state of the switch cabinet once when the current and the voltage change once;
S33, after repeatedly collecting temperature change data of the switch cabinet when the current and the voltage change, carrying out standardization processing on the collected current, voltage and temperature change data, and establishing a relation model of a control instruction and the state parameters of the switch cabinet; the formula for carrying out standardization processing on the collected current, voltage and temperature change data is as follows: Wherein Z represents data before normalization processing,Representing the normalized data; maximum value in state data representing each parameter variable; A minimum value in the state data representing each parameter variable; the calculation formula of the relation model of the control instruction and the state parameter of the switch cabinet is as follows: the relation model between the temperature state parameter in the switch cabinet and the voltage and current set by the control instruction is as follows: Wherein,In order to control the temperature state parameters in the switch cabinet obtained after the instruction execution,The voltage change amount set by the control command is indicated,The current variation set by the control command is indicated,The amount of change in time is indicated,In order to control the temperature state parameters in the switch cabinet collected before the instruction execution,The relation weight of voltage and current is set for the temperature state parameter and the control instruction in the switch cabinet;
S4, analyzing and judging a relation model of the control command and the state parameter of the switch cabinet based on the BP neural network, and determining the relation model of the control command and the state parameter of the switch cabinet; based on BP neural network, analyzing and judging the relation model of the control command and the state parameter of the switch cabinet, and determining the relation model of the control command and the state parameter of the switch cabinet comprises the following steps:
S41, selecting part of initialized historical state parameter data from all initialized historical state parameter data to form a historical state parameter data training set;
S42, determining the structure of the BP neural network, a learning algorithm and selecting related parameters; the BP neural network structure comprises: an input layer, an hidden layer and an output layer; further, the learning algorithm of the BP neural network comprises: setting the input of the input layer asWhereinRepresenting the control instructions and the switch cabinet state parameters in the first historical state parameter data input,The control instruction and the switch cabinet state parameter in the input mth historical state parameter data are represented, and the output of the output layer isWhereinA first predictive switch cabinet temperature state parameter indicative of the output,Representing the temperature state parameter in the output nth prediction switch cabinet, wherein the hidden layer comprises q neurons, v is the weight from the input layer to the hidden layer, and w is the weight from the hidden layer to the output layer; the process of the BP neural network comprises two stages of forward propagation of signals and backward propagation of errors; the forward propagation of the signal is that data is input from an input layer, passes through an hidden layer and finally reaches an output layer; the back propagation of errors is that the data from the output layer to the hidden layer, and finally to the input layer, the weight from the hidden layer to the output layer and the weight from the input layer to the hidden layer are sequentially adjusted; further, the forward propagation of the signal is calculated as: the calculation formula for transmitting the historical state parameter data from the input layer to the hidden layer is as follows: Wherein,Representing the input of the h hidden layer neuron,Representing the weight of the ith input of the input layer to the h hidden layer neuron, and m represents the quantity of input data of the input layer; Representing an input layer ith input; Representing the bias value of the h hidden layer neuron; the calculation formula for transmitting the historical state parameter data from the hidden layer to the output layer is as follows: Wherein,Representing the input of the jth output layer neuron,Representing weights of h hidden layer neurons to j output layer neurons, q representing the number of hidden layer neurons; representing the output of the h hidden layer neuron; Representing the bias value of the jth output layer neuron; further, the output layer directly outputs the received historical state parameter data transmitted by the hidden layer after receiving the historical state parameter data transmitted by the hidden layer; i.e.，Representing the output of the jth output layer neuron; further, calculating an error between the output layer and the expected value, setting an error threshold between the output layer and the expected value, and when the error between the output layer and the expected value is greater than or equal to the set error threshold, sequentially adjusting weights from the hidden layer to the output layer and from the input layer to the hidden layer through a counter propagation stage of the error; the error calculation formula is as follows: Wherein E is an error, and the error is defined as,N is the number of output neurons, which is the expected value of the j-th output; the calculation formula of the weight adjustment is as follows: Wherein,Representing a weight adjustment value, i representing a learning rate, E representing an error, y representing an output of the output layer; further, the weight is regulated in a mode of updating the weight regulating value of the current weight, and the regulated weight is used as a new weight in forward propagation calculation of the signal to carry out iterative calculation until a BP neural network model meeting the error requirement is obtained;
S43, comparing the determined BP neural network with a relation model of the control command and the state parameter of the switch cabinet, analyzing and judging whether the relation model of the control command and the state parameter of the switch cabinet meets the error requirement; respectively inputting a control instruction and a switch cabinet state parameter collected in real time to obtain a BP neural network model meeting the error requirement and a relation model between the control instruction and the switch cabinet state parameter, comparing an output result of the BP neural network model meeting the error requirement with an output result of the relation model between the control instruction and the switch cabinet state parameter, setting an error threshold, and judging whether the output result of the relation model between the control instruction and the switch cabinet state parameter meets the error or not based on the set error threshold; when the comparison results in that the relation model of the control command and the state parameter of the switch cabinet meets the error requirement, setting the relation model of the current control command and the state parameter of the switch cabinet as a prediction model of the relation between the control command and the state parameter of the switch cabinet, and predicting the state parameter in the switch cabinet after the control command for acquiring the state parameter data in the switch cabinet in real time is executed through the prediction model; when the comparison results in that the relation model of the control command and the state parameter of the switch cabinet does not accord with the error requirement, the relation model of the control command and the state parameter of the switch cabinet is reestablished;
s5, after analysis is completed, predicting state parameter data in the switch cabinet after the control instruction is executed through a relation model of the determined control instruction and the state parameter of the switch cabinet, and uploading the predicted state parameter data in the switch cabinet after the control instruction is executed through an IEC61850 communication mode;
S6, intelligently monitoring the running state of the switch cabinet based on state parameter data in the switch cabinet after the predicted control instruction uploaded in the IEC61850 communication mode is executed; the invention also discloses an intelligent monitoring system for the running state of the switch cabinet based on IEC61850 communication, which comprises a data acquisition module, a database, a data processing module, a model construction module, a neural network analysis module, a communication transmission module and a monitoring prediction module; the data acquisition module is used for acquiring the state parameters in the switch cabinet in real time through installed state parameter acquisition equipment; the state parameter acquisition device includes: a current state monitor, a voltage state monitor, and a switchgear temperature state monitor; the current state monitor is used for monitoring the current state of the switch cabinet in real time; the voltage state monitor is used for monitoring the voltage state of the switch cabinet in real time; the switch cabinet temperature state monitor is used for monitoring the temperature state of the switch cabinet in real time; the data processing module is used for filtering and classifying the acquired data; the database is used for storing state parameter data acquired in real time; the model construction module is used for establishing a relation model between the control instruction and the state parameters of the switch cabinet; the neural network analysis module is used for training state parameters in the switch cabinet; the monitoring prediction module is used for monitoring and predicting state parameters in the switch cabinet; the communication transmission module is used for transmitting the monitoring prediction data of the state parameters in the switch cabinet.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.