Processing system and method for detecting environmental information of power material storage

Abstract

A processing system and method for detecting environmental information of electric power material storage, belonging to the technical field of environmental information processing, forms a past queue of environmental information parameters based on sampling values ​​from environmental information sampling pieces, and for each sensor, forms a corresponding past queue of auxiliary information based on sampling values ​​from corresponding sensors, then calculates and obtains the coherence factor of each auxiliary information and environmental information parameter based on such past queues, and obtains the final estimated value of the environmental information parameter at a future time point by combining neural network calculation, and finally transmits the estimated value to a liquid crystal screen for display, thereby not only providing the effect of pre-execution environmental information estimation and pre-notification, but also having better estimation accuracy and pre-notification efficiency, improving the performance of pre-controlling the environmental information of the room where the electric power material storage is located, and being suitable for pre-controlling the change of environmental information in the room where the electric power material storage is located.

Description

Technical Field

[0001] The present invention belongs to the technical field of environmental information processing, and in particular relates to a system and method for processing environmental information for detecting electric power material storage. Background Art

[0002] Various materials and equipment required in the construction and operation of power materials, including cables, transformers, switchgear, transmission towers, optical cables, wire and cable accessories, etc. These power materials are indispensable in the construction and operation of power grids.

[0003] In terms of power material storage management, as mentioned in the prior art solution with patent publication number "CN214540558U", it often includes a refrigeration host, a dehumidifier, a temperature sensor and a humidity sensor connected to the controller. The temperature sensor and the humidity sensor are used to sample the temperature value and the humidity value of the room where the power material storage is located and transmit them to the controller. The controller is used to control the refrigeration host to cool the indoor environment where the power material storage is located based on whether the transmitted temperature value and humidity value of the room where the power material storage is located are higher than the pre-defined temperature critical value and humidity critical value respectively. If the temperature value of the room where the power material storage is located is higher than the pre-defined temperature critical value, the dehumidifier is controlled to dehumidify the indoor environment where the power material storage is located. The temperature value or humidity value of the room where the power material storage is located is the environmental information of the power material storage, and the temperature sensor or the humidity sensor is the environmental information sampling component.

[0004] However, due to the lack of the ability to estimate environmental information in advance and to notify in advance, the ability to grasp the environmental information in the room where the power material storage is located in advance is insufficient, and it is not suitable to grasp the changes in the environmental information in the room where the power material storage is located in advance. Summary of the invention

[0005] In order to solve the defects in the prior art, the present invention proposes a processing system and method for detecting environmental information of electric power material storage, which forms a past queue of environmental information parameters based on the sampling values ​​from the environmental information sampling piece, and for each sensor, forms a corresponding past queue of auxiliary information based on the sampling values ​​from the corresponding sensor, and then calculates and obtains the coherence factor of each auxiliary information and environmental information parameter based on such past queue, and obtains the final estimated value of the environmental information parameter at a future point in time by combining neural network calculation, and finally transmits the estimated value to the liquid crystal screen for display, thereby not only providing the effect of pre-execution of environmental information estimation and pre-notification, but also because the estimation is performed by concentrating the past information of environmental information and the past information of auxiliary information, it has better estimation accuracy and pre-notification efficiency, thereby improving the performance of pre-controlling the environmental information in the room where the electric power material storage is located, and is suitable for pre-controlling the changes in the environmental information in the room where the electric power material storage is located.

[0006] The present invention applies the following technical solutions.

[0007] A method for processing environmental information for detecting power material storage includes:

[0008] The environmental information sampling component samples the environmental information of the room where the power material storage is located in real time, and transmits the sampled value to the controller in real time so that the controller can control the dehumidifier and the refrigeration host; the auxiliary information sampling component samples the auxiliary information of the room where the power material storage is located in real time, and transmits the sampled value to the controller in real time;

[0009] The method for processing environmental information for detecting power material storage also includes:

[0010] Step 1: Based on the sampled values ​​from the environmental information sampling component, a past queue of environmental information parameters is formed, and for each sensor in the auxiliary information sampling component, a past queue of corresponding auxiliary information is formed based on the sampled values ​​from the corresponding sensor;

[0011] Step 2: For each auxiliary information corresponding to each sensor one-to-one, calculate and obtain the coherence factor of the corresponding information and the environmental information parameter according to the corresponding past queue and the past queue of the environmental information parameter;

[0012] Step 3: Based on the past queue of environmental information parameters, use the neural network to estimate in real time the environmental information parameters in the future The estimated value of the sampling time point is 1, where Represents less than The natural number of

[0013] Step 4: For each auxiliary information, according to the corresponding past queue and the past queue of environmental information parameters, use the neural network to estimate the corresponding environmental information parameters in the future. The estimated value of the sampling time point is 2;

[0014] Step 5: Based on environmental information parameters in the future The estimated value 1 and the overall estimated value 2 at each sampling time point are combined with the relevant factors of each auxiliary information and environmental information parameter to obtain the environmental information parameter in the future in real time. The final estimated value at each sampling point;

[0015] Step 6: Parameterize environmental information in the future The final estimated value of each sampling point is instantly transmitted to the LCD screen.

[0016] Furthermore, in Step 1, the past period queue includes sampling time points and the past information corresponding to each sampling time point, Represents a natural number not less than three.

[0017] Furthermore, in Step 1, after forming the past queues of the environmental information parameters and the auxiliary information, the controller also performs a comprehensive movement of the sampling time points in the past queue of the auxiliary information, so that each sampling time point after the movement corresponds one-to-one with each sampling time point in the past queue of the environmental information parameters, and for each sampling time point after the movement, the corresponding and refreshed past queue is obtained according to the following equation: :

[0018] ,

[0019] In the equation, Represents the total amount of time points in the past queue of environmental information parameters, Represents no higher than The natural number of Representatives and The auxiliary information corresponding to the sampling time point after the movement, Representative The time interval from the sampling time point after the movement to the first time point in the past queue of the environmental information parameter, , , and Respectively represent the previous queue of auxiliary information and There is a one-to-one correspondence between adjacent sampling points Auxiliary information, using Bessel third-order fitting algorithm to fit the factors of the third-order function, represents a natural number not less than four, The adjacent sampling time points are The latest sampling time after the movement comes first A sampling point in time.

[0020] Furthermore, Step 2 specifically includes:

[0021] Step 2-1: When faced with an auxiliary information, randomly extract a sub-queue from the corresponding previous queue to form a selected information. , and also extract and select information from the previous queue based on the environmental information parameters. At the same time, another node queue forms a selective information 2 ;

[0022] Step 2-2: Select information Perform the Anderson-Darling test to calculate the total test value , and also select information 2 Perform the Anderson-Darling test to calculate the total test value 2 ;

[0023] Step 2-3: If the total amount of inspection is Total quantity and inspection If all of them are higher than the predefined critical amount, the selection information is obtained according to the following equation: and optional information 2 The correlation factor :

[0024] ,

[0025] In the equation, represents a natural number, Representatives in the selection of information The first Information, Representatives in the selection of information 2 The first Information, Representative Selection Information 1 The mean information within Representative Selection Information 2 The mean of information within

[0026] Step 2-4: The correlation factor It is regarded as the coherence factor of the auxiliary information and environmental information parameters.

[0027] Furthermore, in Step 2-3, the pre-defined critical amount is one twentieth.

[0028] Furthermore, Step 3 specifically includes:

[0029] Step 3-1: Extract information 1 from the past queue of environmental information parameters according to the set method;

[0030] Step 3-2: Send all the extracted information 1 to the vector machine based on the neural network to perform pattern recognition and obtain the environmental information value estimation model 1;

[0031] Step 3-3: Extract the latest one in sequence at a certain point in time from the past queue based on the environmental information parameter The past environmental information of a sampling point in time is sent to an environmental information value estimation model 1, and an estimated value 1 of the environmental information parameter at a future sampling point in time is sent out.

[0032] Further, in Step 3-1, the past queue of environmental information parameters is extracted in sequence at a certain point in time. The past environmental information of the sampling time point is taken as the input information, and the past queue of environmental information parameters is extracted in After the sampling time point and continuously at the time point The past environmental information at a sampling time point is regarded as the sent information 1, and then the input information 1 and the sent information 1 are regarded as the extracted information 1. Here, Represents not less than two and less than The natural number of Represents no higher than A natural number.

[0033] Furthermore, in Step 3-2, the vector machine of the neural network is a support vector machine.

[0034] Furthermore, Step 4 specifically includes:

[0035] Step 4-1: For each auxiliary information, extract the corresponding extracted information 2 from the corresponding previous queue according to the following method:

[0036] The past auxiliary information of the sampling time points that are sequentially continuous in time is extracted from the corresponding past queue as the input information 2, and the past auxiliary information of the sampling time points that are sequentially continuous in time is extracted from the past queue of the environmental information parameter as the input information 2. The past environmental information of the sampling time points after the sampling time points and continuously in time points is regarded as the sent information 2, and then the sent information 2 and the sent information 2 are regarded as the extracted information 2. Here, Represents not less than two and less than The natural number of Represents no higher than The natural number of

[0037] Step 4-2: For each auxiliary information, use the corresponding total extracted information 2 to perform pattern recognition on the vector machine based on the neural network to obtain the environmental information value estimation model 2;

[0038] Step 4-3: For each auxiliary information, extract the latest one in sequence at a certain point in time from the corresponding past queue. The auxiliary information of the past sampling time point is sent to the environmental information value estimation model 2, and the corresponding environmental information parameters are sent out in the future. The estimated value of the sampling time point is 2.

[0039] Further, in Step 5, facing the future At each sampling time point in the year, the final estimated value of the environmental information parameter at the corresponding time point is obtained according to the following equation:

[0040] ,

[0041] In the equation, Represents no higher than The natural number of Represents environmental information parameters in the future The sampling time point The final estimated value at a point in time, represents the total amount of auxiliary information, Represents no higher than The natural number of Representatives and The criticality factor corresponding to the auxiliary information is Representatives and The coherence factor corresponding to the auxiliary information is Represents no higher than The natural number of Representatives and The coherence factor corresponding to the auxiliary information is Represents the environmental information parameter in The estimated value at a point in time is Representatives and The auxiliary information corresponds to the environmental information parameter in the The estimated value at a point in time is 2.

[0042] A processing system for detecting environmental information of power material storage, comprising:

[0043] A refrigeration host, a dehumidifier, an environmental information sampling component, an auxiliary information sampling component, a controller and a liquid crystal display screen. Here, the refrigeration host, the dehumidifier, the environmental information sampling component, the auxiliary information sampling component and the liquid crystal display screen are all connected to the controller;

[0044] Environmental information sampling device, used to instantly sample the indoor environmental information where the power materials are stored, and instantly transmit the sampled value to the controller so that the controller can control the dehumidifier and refrigeration host;

[0045] Auxiliary information sampling piece, used to instantly sample the auxiliary information in the room where the power materials are stored, and instantly transmit the sampled value to the controller;

[0046] The processing system for detecting environmental information of power material storage also includes:

[0047] A queue module, which is used to form a past queue of environmental information parameters based on the sampled values ​​from the environmental information sampling element, and for each sensor in the auxiliary information sampling element, form a past queue of corresponding auxiliary information based on the sampled values ​​from the corresponding sensor;

[0048] A coherence module, which is used to face each auxiliary information corresponding to each sensor one-to-one, and calculate the coherence factor of the corresponding information and the environmental information parameter according to the corresponding past queue and the past queue of the environmental information parameter;

[0049] The inference module is used to use the neural network to estimate the future environmental information parameters in real time based on the past queue of environmental information parameters. The estimated value of the sampling time point is 1, where Represents less than The natural number of

[0050] The estimation module is used to face each auxiliary information, based on the corresponding past queue and the past queue of environmental information parameters, use the neural network to instantly estimate the corresponding and environmental information parameters in the future The estimated value of the sampling time point is 2;

[0051] The calculation module is used to calculate the future The estimated value 1 and the overall estimated value 2 at each sampling time point are combined with the relevant factors of each auxiliary information and environmental information parameter to obtain the environmental information parameter in the future in real time. The final estimated value at each sampling point;

[0052] The transmission module is used to transmit the environmental information parameters in the future The final estimated value of each sampling point is instantly transmitted to the LCD screen.

[0053] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include:

[0054] Refrigeration host, dehumidifier, environmental information sampling component, auxiliary information sampling component, controller and LCD screen, here, the controller is used to form a past queue of environmental information parameters according to the sampling value from the environmental information sampling component, and for each sensor, form a corresponding past queue of auxiliary information according to the sampling value from the corresponding sensor, then calculate and obtain the coherence factor of each auxiliary information and environmental information parameter according to such past queue, and combine the neural network calculation to obtain the final estimated value of the environmental information parameter at a future point in time, and finally the estimated value is transmitted to the LCD screen for display, thereby not only providing the effect of pre-execution of environmental information estimation and pre-notification, but also because the past information of environmental information and the past information of auxiliary information are concentrated to perform estimation, it has better estimation accuracy and pre-notification efficiency, improves the performance of pre-controlling the environmental information in the room where the electric power material storage is located, and is suitable for pre-controlling the changes in the environmental information in the room where the electric power material storage is located. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a partial flow chart of the processing method for detecting environmental information of power material storage described in the present invention;

[0056] Figure 2 It is a partial structural diagram of the processing system for detecting environmental information of power material storage described in the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely expressed in combination with the drawings in the embodiments of the present invention. The embodiments expressed in this application are only some embodiments of the present invention, not all embodiments. According to the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the protection scope of the present invention.

[0058] like Figure 1 As shown, a processing method for detecting environmental information of power material storage according to the present invention includes:

[0059] The environmental information sampling component samples the environmental information of the room where the power material storage is located in real time, and transmits the sampled value to the controller in real time so that the controller can control the dehumidifier and the refrigeration host; the auxiliary information sampling component samples the auxiliary information of the room where the power material storage is located in real time, and transmits the sampled value to the controller in real time;

[0060] The processing method for detecting environmental information of power material storage runs on the controller and also includes:

[0061] Step 1: Based on the sampled values ​​from the environmental information sampling component, a past queue of environmental information parameters is formed, and for each sensor in the auxiliary information sampling component, a past queue of corresponding auxiliary information is formed based on the sampled values ​​from the corresponding sensor;

[0062] In a preferred but non-limiting embodiment of the present invention, in Step 1, the past period queue includes sampling time points and the past information corresponding to each sampling time point (the past information of the past queue is: if the past queue is a past queue of environmental information parameters, the past information of the past queue is the environmental information of the room where the power material storage is located in the past; if the past queue is a past queue of auxiliary information, the past information of the past queue is the auxiliary information of the room where the power material storage is located in the past), Represents a natural number not less than three.

[0063] In Step 1, since the environmental information sampling element and the auxiliary information sampling element are information corresponding to the real-time sampling according to a predetermined time interval, the sampling values ​​can be registered in sequence according to the order of the sampling time points, thereby forming a past queue of different parameters. The sampling time points can be sampling time points every 1s, every 2s, or every 3s, etc., that is, the past queue of the environmental information parameters carries the environmental information before the present. The past information of the sampling time point before the present The past information of the sampling time point before the present time ... the past information of the second sampling time point before the present time and the past information of the first sampling time point before the present time, etc.; the past queue of auxiliary information includes the auxiliary information before the present time. The past information of the sampling time point before the present The past information at the sampling time point before the present time...the past information at the second sampling time point before the present time and the past information at the first sampling time point before the present time, etc.

[0064] Because the environmental information sampling components and the auxiliary information sampling components have inaccurate sampling time points, according to the characteristics that the adjacent sampling time points of the environmental information parameters and the auxiliary information have the same time interval (because the information is sampled at regular intervals, such as sampling every 3 seconds or less), in order to better maintain the coordination of the environmental information parameters and the auxiliary information past queues at the sampling time points, thereby preventing the above sampling time point inaccuracy from being detrimental to the accuracy of subsequent information analysis, in a preferred but non-restrictive embodiment of the present invention, in Step 1, after forming the past queues of the environmental information parameters and the auxiliary information, the controller also performs a comprehensive movement of the sampling time points in the past queues of the auxiliary information, so that each sampling time point after the movement corresponds one-to-one with each sampling time point in the past queue of the environmental information parameters, and for each sampling time point after the movement, the corresponding and refreshed past queue is obtained according to the following equation :

[0065] ,

[0066] In the equation, Represents the total amount of time points in the past queue of environmental information parameters, Represents no higher than The natural number of Representatives and The auxiliary information corresponding to the sampling time point after the movement, Representative The time interval from the sampling time point after the movement to the first time point in the past queue of the environmental information parameter (the time point of this application is the sampling time point), , , and Respectively represent the previous queue of auxiliary information and There is a one-to-one correspondence between adjacent sampling points Auxiliary information, using Bessel third-order fitting algorithm to fit the factors of the third-order function, represents a natural number not less than four, The adjacent sampling time points are The latest sampling time after the movement comes first Sampling time points. Just as if each sampling time point in the past queue of the environmental information parameter is one minute past nine, two minutes past nine... four minutes past nine, fifteen minutes past nine, etc., and each sampling time point in the past queue of the auxiliary information is twenty-two seconds past nine, one minute and twenty-two seconds past nine... thirteen minutes and twenty-two seconds past nine, fourteen minutes and twenty-two seconds past nine, etc., then after all such sampling time points are moved backward for thirty-eight seconds, each sampling time point after the movement will be one minute past nine, two minutes past nine... four minutes past nine, fifteen minutes past nine, etc., and correspond one-to-one and align with each sampling time point in the past queue of the environmental information parameter.

[0067] Step 2: For each auxiliary information corresponding to each sensor one-to-one, calculate and obtain the coherence factor of the corresponding information and the environmental information parameter according to the corresponding past queue and the past queue of the environmental information parameter;

[0068] In Step 2, the degree of effect of different auxiliary information on the environmental information parameters is definitely different, so it is necessary to calculate and obtain the coherence factor of each auxiliary information and the environmental information parameter for subsequent centralized estimation of the environmental information, and to ensure the accuracy of the centralized estimation value, that is, facing each auxiliary information corresponding to each sensor one-to-one, according to the corresponding past queue and the past queue of the environmental information parameter, calculate and obtain the coherence factor of the corresponding information and the environmental information parameter. In a preferred but non-limiting embodiment of the present invention, Step 2 specifically includes:

[0069] Step 2-1: When faced with an auxiliary information, randomly extract a sub-queue from the corresponding previous queue to form a selected information. , and also extract and select information from the previous queue based on the environmental information parameters. Another node queue in the same period (same sampling period) forms the selected information 2 ;

[0070] In Step 2-1, if the auxiliary information is the wind speed value, the sampling time points from 12:00 on July 6 to 18:00 on July 6 and the corresponding sampling values ​​can be extracted from the past queue of wind speed values ​​to form the selected information. At the same time, the sampling time points from 12:00 on July 6 to 18:00 on July 6 and the corresponding sampling values ​​are extracted from the past queue of the environmental information parameter to form the selected information 2. .

[0071] Step 2-2: Select information Perform the Anderson-Darling test to calculate the total test value , and also select information 2 Perform the Anderson-Darling test to calculate the total test value 2 ;

[0072] Step 2-3: If the total amount of inspection is Total quantity and inspection If all of them are higher than the predefined critical amount, the selection information is obtained according to the following equation: and optional information 2 The correlation factor :

[0073] ,

[0074] In the equation, represents a natural number, Representatives in the selection of information The first Information, Representatives in the selection of information 2 The first Information, Representative Selection Information 1 The mean information within Representative Selection Information 2 The mean of information within

[0075] In a preferred but non-limiting embodiment of the present invention, in Step 2-3, the pre-defined critical amount is one twentieth.

[0076] Step 2-4: The correlation factor As the coherence factor of the auxiliary information and environmental information parameters as described in step 2-1.

[0077] Step 3: Based on the past queue of environmental information parameters, use the neural network to estimate in real time the environmental information parameters in the future The estimated value of the sampling time point is 1, where Represents less than The natural number of

[0078] In a preferred but non-limiting embodiment of the present invention, Step 3 specifically includes: based on the past queue of environmental information parameters, using a neural network to instantly estimate the environmental information parameters in the future The estimated value of the sampling time point is:

[0079] Step 3-1: Extract information 1 from the past queue of environmental information parameters according to the set method;

[0080] In a preferred but non-limiting embodiment of the present invention, in Step 3-1, the past period queue of the environmental information parameter is extracted in sequence at a time point. The past environmental information of the sampling time point is taken as the input information, and the past queue of environmental information parameters is extracted in After the sampling time point and continuously at the time point The past environmental information at a sampling time point is regarded as the sent information 1, and then the input information 1 and the sent information 1 are regarded as the extracted information 1. Here, Represents not less than two and less than The natural number of Represents no higher than A natural number.

[0081] Just as the sampling time point is every 60s, It can be twenty. It can be ten, so that the past environmental information from 10:05 to 10:02 minutes (that is, a total of 20 sampling time points) can be extracted from the past queue of the environmental information parameters and used as input information one, and the past environmental information from 10:02 to 10:034 minutes (that is, a total of 10 sampling time points) can be extracted as output information one, thereby obtaining an extracted information one.

[0082] Step 3-2: Send all the extracted information 1 to the vector machine based on the neural network to perform pattern recognition and obtain the environmental information value estimation model 1;

[0083] In a preferred but non-limiting embodiment of the present invention, in Step 3-2, the vector machine of the neural network is a support vector machine.

[0084] Step 3-3: Extract the latest one in sequence at a certain point in time from the past queue based on the environmental information parameter The past environmental information of a sampling point in time is sent to an environmental information value estimation model 1, and an estimated value 1 of the environmental information parameter at a future sampling point in time is sent out.

[0085] In Step 3-3, if the current time is 3:00 p.m., the past environmental information from 4:41 p.m. to 3:00 p.m. (that is, a total of 20 sampling points) can be extracted from the past queue of environmental information parameters and sent to environmental information value estimation model 1 to obtain the estimated value 1 of the environmental information parameters from 1 minute past 3:00 p.m. to 10 minutes past 3:00 p.m. (that is, a total of 10 sampling points).

[0086] Step 4: For each auxiliary information, according to the corresponding past queue and the past queue of environmental information parameters, use the neural network to estimate the corresponding environmental information parameters in the future. The estimated value of the sampling time point is two; the neural network can be an autoencoder.

[0087] In a preferred but non-limiting embodiment of the present invention, Step 4 specifically includes:

[0088] Step 4-1: For each auxiliary information, extract the corresponding extracted information 2 from the corresponding previous queue according to the following method:

[0089] The past auxiliary information of the sampling time points that are sequentially continuous in time is extracted from the corresponding past queue as the input information 2, and the past auxiliary information of the sampling time points that are sequentially continuous in time is extracted from the past queue of the environmental information parameter as the input information 2. The past environmental information of the sampling time points after the sampling time points and continuously in time points is regarded as the sent information 2, and then the sent information 2 and the sent information 2 are regarded as the extracted information 2. Here, Represents not less than two and less than The natural number of Represents no higher than The natural number of

[0090] Step 4-2: For each auxiliary information, use the corresponding total extracted information to perform pattern recognition on a vector machine based on a neural network to obtain an environmental information value estimation model 2; the vector machine can be a support vector machine.

[0091] Step 4-3: For each auxiliary information, extract the latest one in sequence at a certain point in time from the corresponding past queue. The auxiliary information of the past sampling time point is sent to the environmental information value estimation model 2, and the corresponding environmental information parameters are sent out in the future. The estimated value of the sampling time point is 2.

[0092] Step 5: Based on environmental information parameters in the future The estimated value 1 and the overall estimated value 2 at each sampling time point are combined with the relevant factors of each auxiliary information and environmental information parameter to obtain the environmental information parameter in the future in real time. The final estimated value at each sampling point;

[0093] In Step 5, since the correlation factor is proportional to the criticality factor of the auxiliary information on the environmental information parameter (that is, the higher the correlation factor, the higher the criticality factor), the correlation factor of each auxiliary information and environmental information parameter can be combined to achieve the centralized estimation of environmental information, that is, based on the environmental information parameter in the future The estimated value 1 and the overall estimated value 2 at each sampling time point are combined with the relevant factors of each auxiliary information and environmental information parameter to obtain the environmental information parameter in the future in real time. The final estimated value at the sampling time point.

[0094] In a preferred but non-limiting embodiment of the present invention, in Step 5, facing the future At each sampling time point in the year, the final estimated value of the environmental information parameter at the corresponding time point is obtained according to the following equation:

[0095] ,

[0096] In the equation, Represents no higher than The natural number of Represents environmental information parameters in the future The sampling time point The final estimated value at a point in time, represents the total amount of auxiliary information, Represents no higher than The natural number of Representatives and The criticality factor corresponding to the auxiliary information is Representatives and The coherence factor corresponding to the auxiliary information is Represents no higher than The natural number of Representatives and The coherence factor corresponding to the auxiliary information is Represents the environmental information parameter in The estimated value at a point in time is Representatives and The auxiliary information corresponds to the environmental information parameter in the The estimated value at a point in time is 2.

[0097] Step 6: Parameterize environmental information in the future The final estimated value of each sampling point is transmitted to the LCD screen in real time. This can improve the ability to predict the environmental information in the room where the power material storage is located, and is suitable for predicting the changes in the environmental information in the room where the power material storage is located.

[0098] like Figure 2 As shown, the processing system for detecting environmental information of power material storage according to the present invention includes:

[0099] A refrigeration host, a dehumidifier, an environmental information sampling component, an auxiliary information sampling component, a controller and a liquid crystal display screen. Here, the refrigeration host, the dehumidifier, the environmental information sampling component, the auxiliary information sampling component and the liquid crystal display screen are all connected to the controller;

[0100] The environmental information sampling component is used to instantly sample the environmental information of the room where the electric power material warehouse is located, and instantly transmit the sampled value (the sampled value is the environmental information of the room where the electric power material warehouse is located) to the controller so that the controller can control the dehumidifier and the refrigeration host. Specifically, the temperature sensor and the humidity sensor are used to respectively sample the temperature value and the humidity value of the room where the electric power material warehouse is located and transmit them to the controller. The controller is used to control the refrigeration host to cool the indoor environment of the electric power material warehouse if the temperature value of the room where the electric power material warehouse is located is higher than the pre-defined temperature critical value, if the humidity value of the room where the electric power material warehouse is located is higher than the pre-defined humidity critical value, then control the dehumidifier to dehumidify the indoor environment of the electric power material warehouse. The temperature value or humidity value of the room where the electric power material warehouse is located is the environmental information of the electric power material warehouse, and the temperature sensor or the humidity sensor is the environmental information sampling component.

[0101] The auxiliary information sampling component is used to instantly sample the auxiliary information in the room where the electric power material storage is located, and instantly transmit the sampled value (the sampled value is the auxiliary information in the room where the electric power material storage is located) to the controller; the auxiliary information sampling component includes a wind speed sensor and a wind force sensor, and the wind speed sensor and the wind sensor are used to respectively sample the wind speed value and wind force value in the room where the electric power material storage is located, and the wind speed value or wind force value in the room where the electric power material storage is located is the auxiliary information.

[0102] The processing system for detecting environmental information of power material storage also includes:

[0103] A queue module, which is used to form a past queue of environmental information parameters based on the sampled values ​​from the environmental information sampling element, and for each sensor in the auxiliary information sampling element, form a past queue of corresponding auxiliary information based on the sampled values ​​from the corresponding sensor;

[0104] A coherence module, which is used to face each auxiliary information corresponding to each sensor one-to-one, and calculate the coherence factor of the corresponding information and the environmental information parameter according to the corresponding past queue and the past queue of the environmental information parameter;

[0105] The inference module is used to use the neural network to estimate the future environmental information parameters in real time based on the past queue of environmental information parameters. The estimated value of the sampling time point is 1, where Represents less than The natural number of

[0106] The estimation module is used to face each auxiliary information, based on the corresponding past queue and the past queue of environmental information parameters, use the neural network to instantly estimate the corresponding and environmental information parameters in the future The estimated value of the sampling time point is 2;

[0107] The calculation module is used to calculate the future The estimated value 1 and the overall estimated value 2 at each sampling time point are combined with the relevant factors of each auxiliary information and environmental information parameter to obtain the environmental information parameter in the future in real time. The final estimated value at each sampling point;

[0108] The transmission module is used to transmit the environmental information parameters in the future The final estimated value of each sampling point is instantly transmitted to the LCD screen.

[0109] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include:

[0110] Refrigeration host, dehumidifier, environmental information sampling component, auxiliary information sampling component, controller and LCD screen, here, the controller is used to form a past queue of environmental information parameters according to the sampling value from the environmental information sampling component, and for each sensor, form a corresponding past queue of auxiliary information according to the sampling value from the corresponding sensor, then calculate and obtain the coherence factor of each auxiliary information and environmental information parameter according to such past queue, and combine the neural network calculation to obtain the final estimated value of the environmental information parameter at a future point in time, and finally the estimated value is transmitted to the LCD screen for display, thereby not only providing the effect of pre-execution of environmental information estimation and pre-notification, but also because the past information of environmental information and the past information of auxiliary information are concentrated to perform estimation, it has better estimation accuracy and pre-notification efficiency, improves the performance of pre-controlling the environmental information in the room where the electric power material storage is located, and is suitable for pre-controlling the changes in the environmental information in the room where the electric power material storage is located.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not deviate from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for processing environmental information for detecting power material storage, characterized in that: include: The environmental information sampling component samples the environmental information of the room where the power material storage is located in real time, and transmits the sampled value to the controller in real time so that the controller can control the dehumidifier and the refrigeration host; the auxiliary information sampling component samples the auxiliary information of the room where the power material storage is located in real time, and transmits the sampled value to the controller in real time; The method for processing environmental information for detecting power material storage also includes: Step 1: Form a past queue of environmental information parameters based on the sampled values ​​from the environmental information sampling component, and for each sensor in the auxiliary information sampling component, form a past queue of corresponding auxiliary information based on the sampled values ​​from the corresponding sensor; Step 2: For each auxiliary information corresponding to each sensor one-to-one, calculate and obtain the coherence factor of the corresponding information and the environmental information parameter according to the corresponding past queue and the past queue of the environmental information parameter; Step 3: Based on the past queue of environmental information parameters, use the neural network to estimate in real time the environmental information parameters in the future The estimated value of the sampling time point is 1, where Represents less than The natural number of represents a natural number not less than three; Step 4: For each auxiliary information, based on the corresponding past queue and the past queue of environmental information parameters, use the neural network to instantly estimate the corresponding environmental information parameters in the future. The estimated value of the sampling time point is 2; Step 5: Based on environmental information parameters in the future The estimated value 1 and the overall estimated value 2 at each sampling time point are combined with the relevant factors of each auxiliary information and environmental information parameter to obtain the environmental information parameter in the future in real time. The final estimated value at each sampling point; Step 6: Parameterize environmental information in the future The final estimated value of each sampling point is instantly transmitted to the LCD screen; In step 5, facing the future At each sampling time point within the sampling time points, the final estimated value of the environmental information parameter at the corresponding time point is obtained according to the following equation: In the equation, Represents no higher than The natural number of Represents environmental information parameters in the future The sampling time point The final estimated value at a point in time, represents the total amount of auxiliary information, Represents no higher than The natural number of Representatives and The criticality factor corresponding to the auxiliary information is Representatives and The coherence factor corresponding to the auxiliary information is Represents no higher than The natural number of Representatives and The coherence factor corresponding to the auxiliary information is Represents the environmental information parameter in The estimated value at a point in time is Representatives and The auxiliary information corresponds to the environmental information parameter in the The estimated value at a point in time is 2.

2. The method for processing environmental information for detecting power material storage according to claim 1 is characterized in that: In step 1, the past queue includes sampling time points and the past information corresponding to each sampling time point.

3. The method for processing environmental information for detecting power material storage according to claim 2 is characterized in that: In step 1, after forming the past queues of environmental information parameters and auxiliary information, the controller also performs a comprehensive movement of the sampling time points in the past queue of auxiliary information to make each moved sampling time point correspond one-to-one with each sampling time point in the past queue of environmental information parameters, and for each moved sampling time point, the corresponding and refreshed past queue is obtained according to the following equation: : In the equation, Represents the total amount of time points in the past queue of environmental information parameters, Represents no higher than The natural number of Representatives and The auxiliary information corresponding to the sampling time point after the movement, Representative The time interval from the sampling time point after the movement to the first time point in the past queue of the environmental information parameter, , , and Respectively represent the previous queue of auxiliary information and There is a one-to-one correspondence between adjacent sampling points Auxiliary information, using Bessel third-order fitting algorithm to fit the factors of the third-order function, represents a natural number not less than four, The adjacent sampling time points are The latest sampling time after the movement comes first A sampling point in time.

4. The method for processing environmental information for detecting power material storage according to claim 3 is characterized in that: Step 2 specifically includes: Step 2-1: When faced with an auxiliary message, randomly extract a sub-queue from the corresponding previous queue to form a selected message. , and also extract and select information from the previous queue based on the environmental information parameters. At the same time, another node queue forms a selective information 2 ; Step 2-2: Select the information Perform the Anderson-Darling test to calculate the total test value , and also select information 2 Perform the Anderson-Darling test to calculate the total test value 2 ; Step 2-3: If the total amount of inspection is Total quantity of inspection If all of them are higher than the predefined critical amount, the selection information is obtained according to the following equation: and optional information 2 The correlation factor : In the equation, represents a natural number, Representatives in the selection of information The first Information, Representatives in the selection of information 2 The first Information, Representative Selection Information 1 The mean information within Representative Selection Information 2 The mean of information within Step 2-4: Add the correlation factor As the coherence factor of the auxiliary information and environmental information parameters; In step 2-3, the predefined critical amount is one twentieth.

5. The method for processing environmental information for detecting power material storage according to claim 4 is characterized in that: Step 3 specifically includes: Step 3-1: Extract according to the set method to obtain extracted information 1; Step 3-2: Send all the extracted information 1 to a vector machine based on a neural network to perform pattern recognition, and obtain an environmental information value estimation model 1; Step 3-3: Extract the latest consecutive data from the past queue according to the environmental information parameter The past environmental information of the sampling time point is sent to the environmental information value estimation model 1, and the environmental information parameters are obtained in the future. The estimated value at a sampling point in time is one.

6. The method for processing environmental information for detecting power material storage according to claim 5 is characterized in that: In step 3-1, the previous period queue of the environmental information parameter is extracted in sequence at a certain point in time. The past environmental information of the sampling time point is taken as the input information, and the past queue of environmental information parameters is extracted in After the sampling time point and continuously at the time point The past environmental information at a sampling time point is regarded as the sent information 1, and then the input information 1 and the sent information 1 are regarded as the extracted information 1. Here, Represents not less than two and less than The natural number of Represents no higher than A natural number.

7. The method for processing environmental information for detecting power material storage according to claim 6 is characterized in that: In step 3-2, the vector machine of the neural network is a support vector machine.

8. The method for processing environmental information for detecting power material storage according to claim 7 is characterized in that: Step 4 specifically includes: Step 4-1: For each auxiliary information, extract the corresponding extracted information 2 from the corresponding previous queue according to the following method: Extract the corresponding past period queues that are continuous at the time point The previous auxiliary information of the sampling time point is taken as the input information 2, and the previous queue of the environmental information parameter is extracted in After sampling time points and continuously at the time points The past environmental information at a sampling time point is taken as the sent information 2, and then the sent information 2 and the sent information 2 are taken as the extracted information 2. Here, Represents not less than two and less than The natural number of Represents no higher than The natural number of Step 4-2: For each auxiliary information, use the corresponding whole extracted information 2 to perform pattern recognition on the vector machine based on the neural network to obtain the environmental information value estimation model 2; Step 4-3: For each auxiliary information, extract the latest one that continues in sequence at a certain point in time from the corresponding past queue. The auxiliary information of the past sampling time point is sent to the environmental information value estimation model 2, and the corresponding environmental information parameters are sent out in the future. The estimated value of the sampling time point is 2.

Images