Internet of Things information management and control platform and method

Abstract

The invention discloses an Internet of Things information management and control platform and method, which are used for calculating the outlier attribute quantity of the Internet of Things information of goods, increasing the distinction degree between the outlier information and reasonable Internet of Things information, and improving the identification function of the outlier information. The method comprises the following steps of: cutting an Internet of Things information attribute vector of goods in an Internet of Things information group of the goods into different groups by using a grouping method, extending an identification space of outlier information, calculating to obtain an Internet of Things information outlier parameter of the goods, and when the Internet of Things information outlier parameter of the goods is higher than an Internet of Things information outlier critical quantity of the goods, carrying out outlier identification on the goods. According to the method and the device, the message for informing the harsh environment of the goods is displayed on the client, so that the application function of the Internet of Things information of the previous goods is improved, the extraction function of the Internet of Things information attributes of the goods is improved, and the sensitivity and the accuracy of outlier detection of the Internet of Things information of the goods are improved.

Description

Technical Field

[0001] The present invention belongs to the field of Internet of Things information technology, and specifically relates to an Internet of Things information management and control platform and method. Background Art

[0002] The Internet of Things (IoT) technology originated from the field of media. IoT refers to connecting any object to the network through information sensing devices according to agreed protocols, and objects exchange and communicate information through information dissemination media to achieve intelligent identification, positioning, tracking, supervision and other functions.

[0003] To implement control over IoT information, it is often necessary to use the existing technical solution with patent publication number "CN108830529A", that is, the temperature sensor, humidity sensor, pressure sensor, and client are all connected to the server for communication. The temperature sensor, humidity sensor, and pressure sensor are used to sample the temperature, humidity, and pressure values ​​of the environment in which the goods are located, respectively, and transmit the sampled temperature, humidity, and pressure values ​​of the environment in which the goods are located to the server. The server forwards the temperature, humidity, and pressure values ​​of the environment in which the goods are located to the client based on the collected temperature, humidity, and pressure values ​​of the environment in which the goods are located. The temperature, humidity, and pressure values ​​of the environment in which the goods are located are the IoT information of the goods, thereby achieving the purpose of controlling the IoT information of the goods.

[0004] Among the IoT information of goods forwarded by the server to the client, even if the IoT information of goods of different categories are not independent and change in the same way, the IoT information of goods has very close correlation. At present, when the client often uses the CLARANS method to identify outlier IoT information, it often analyzes and groups the IoT information of goods of different categories separately, and ignores the correlation between information of different categories, which makes the attribute extraction function of IoT information of goods poor. There are also some correlation conditions between the synchronized IoT information of goods. When only the real-time IoT information of goods is detected, the sensitivity and accuracy of outlier detection are not high, and it is not easy to generate the initial notification of outliers in the IoT information of goods. Summary of the invention

[0005] In order to solve the defects existing in the prior art, the present invention proposes an Internet of Things information management and control platform and method, which calculates the outlier attribute quantity of the Internet of Things information of goods, increases the distinction between outlier information and reasonable Internet of Things information, and improves the recognition function of outlier information; uses a grouping method to cut the Internet of Things information attribute vectors of goods in the Internet of Things information group of goods into different groups, extends the recognition space of outlier information, calculates and obtains the outlier parameters of the Internet of Things information of goods, and when the outlier parameters of the Internet of Things information of goods are higher than the critical amount of the outlier of the Internet of Things information of goods, displays a message on the client to notify that the environment in which the goods are located is bad, improves the application function of the Internet of Things information of past goods, and improves the extraction function of the Internet of Things information attributes of goods, thereby improving the sensitivity and accuracy of outlier detection of the Internet of Things information of goods.

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

[0007] An Internet of Things information management and control method, comprising:

[0008] The temperature sensor, humidity sensor, and pressure sensor respectively sample the temperature value, humidity value, and air pressure value of the environment where the goods are located, and transmit the sampled temperature value, humidity value, and air pressure value of the environment where the goods are located to the server. The server performs processing based on the temperature value, humidity value, and air pressure value of the environment where the goods are located, and forwards the temperature value, humidity value, and air pressure value of the environment where the goods are located to the client. The temperature value, humidity value, and air pressure value of the environment where the goods are located is the IoT information of the goods;

[0009] The method for the client to process the IoT information of the goods includes:

[0010] S1: Obtain IoT information of each category of goods from the server and cut it into IoT information segments of multiple goods;

[0011] S2: According to the outlier amplitude of the random change and scale change of the IoT information of different goods in the IoT information segment of each product when the outlier occurs, the IoT information attribute vector of the goods in the IoT information segment of each product is determined;

[0012] S3: according to the IoT information attribute vector interval of the goods between the IoT information segments of all goods in the IoT information group of the goods, determine the IoT information outlier attribute quantity of the goods in the IoT information group of the goods;

[0013] S4: according to the grouping arrangement of the IoT information attribute vectors of all the IoT information segments of the goods in the IoT information group of the goods and the IoT information outlier attribute quantity of the goods, determine the IoT information outlier parameter of the goods in the IoT information group of the goods;

[0014] S5: Detect the real-time IoT information group of the goods based on the IoT information outlier parameters of the goods.

[0015] Preferably, the sampling rate of the temperature sensor, humidity sensor and pressure sensor for IoT information is g, where g is 400 times per second.

[0016] Preferably, in S1, the same grouping processing is performed on the obtained IoT information of the three categories of goods, and each segment formed by the IoT information of the goods is defined as the IoT information segment of the goods. The corresponding IoT information segments of the goods of each category are respectively defined as: air pressure value segment, temperature value segment and humidity value segment. Here, the number of IoT information contained in the IoT information segment of each goods is P.

[0017] Preferably, S2 specifically includes:

[0018] S2-1: According to the random outlier distribution of air pressure values ​​in a short period of time, determine the IoT information fluctuation attribute quantity of the IoT information segment of each item;

[0019] S2-2: Determine the IoT information scale attribute quantity of the goods in the IoT information segment of each goods based on the corresponding change amplitude between the maximum scale differences of the humidity and temperature values.

[0020] Preferably, S2-1 specifically includes: using an autoencoder to clean up outliers in the air pressure value to obtain air pressure benchmark Internet of Things information.

[0021] Preferably, S2-1 specifically further includes: defining the interval between the air pressure value at the corresponding time point and the air pressure reference IoT information as the air pressure interval;

[0022] The quotient obtained by dividing the variance of the air pressure interval in the Internet of Things information segment of each commodity by the maximum amount of the air pressure interval is defined as the Internet of Things information fluctuation attribute amount of the commodity in the Internet of Things information segment of each commodity.

[0023] Preferably, S2-2 specifically includes: using an autoencoder to process the temperature value to obtain temperature reference IoT information, using a direct method to find out all maximum values ​​of the temperature reference IoT information, and defining the IoT information sequence code corresponding to each maximum value as a temperature maximum value scaling sequence code, and using the same processing method as the temperature value to process the humidity value to obtain humidity reference IoT information and humidity maximum value scaling sequence code;

[0024] The minimum number of maximum values ​​in the temperature reference IoT information and the humidity reference IoT information is defined as the maximum reference number s, and the modulus of the amount obtained by subtracting the corresponding temperature maximum value scaling order code from the humidity maximum value scaling order code according to the maximum reference number is defined as the scaling difference of each IoT information;

[0025] The sum of the quotients obtained by dividing the IoT information scaling differences of all goods in the IoT information segment of each goods by the IoT information scaling difference of the lowest goods is defined as the IoT information scaling attribute quantity of the goods in the IoT information segment of each goods.

[0026] Preferably, S3 specifically includes: using the IoT information segment of the current goods as the starting segment, then taking the IoT information segments of N-1 goods backward to form an IoT information group of the goods of the starting segment, each IoT information group of the goods contains a total of N IoT information segments of the goods, and defining the information entropy between the starting segment and the IoT information segments of other goods in the IoT information group of the goods in which it is located as the inter-segment spacing;

[0027] The outlier attribute amount of the IoT information of the goods in the IoT information group of the goods is calculated, which is:

[0028] In the IoT information group of the goods where the starting segment is located, the attribute vector arrangement criticality between the starting segment and each other IoT information segment of the goods in the IoT information group where the starting segment is located is calculated according to the segment spacing between the starting segment and each other IoT information segment of the goods in the IoT information group where the starting segment is located and the IoT information attribute vector spacing of the goods;

[0029] Sum the attribute vector arrangement criticalities of all other goods' IoT information segments in the goods' IoT information group, perform forward aggregation on the sum value and the IoT information attribute vector of the starting segment, and define the forward aggregation value as the attribute vector improvement amount of the starting segment.

[0030] Preferably, S4 specifically includes: grouping the IoT information attribute vectors of the goods in the IoT information segments of all goods in the IoT information group of the goods using the CLARANS method to obtain multiple groups and group centroids;

[0031] The number of elements in each group in the IoT information group of a random commodity and the centroid of the group are combined as the IoT information outlier attribute quantity of the IoT information group of the commodity in the starting segment, and the IoT information outlier parameter of the IoT information group of the random commodity is determined, which is:

[0032] The number of IoT information in each group in the IoT information group of the goods is taken as the outlier attribute of the IoT information of the goods in the IoT information group of the goods at the starting segment, and the value of the forward aggregation is defined as the group outlier value of each group in the IoT information group of the goods;

[0033] Next, the sum of the group outlier values ​​of all groups in the IoT information group of the goods is calculated to obtain the IoT information outlier parameters of the goods in the IoT information group of the goods.

[0034] Preferably, S5 specifically includes: obtaining N reasonable IoT information groups of goods in the IoT information of past goods, taking the centroid of all groups of the IoT information groups of each goods as the highest value in the IoT information outlier attribute value of the goods in the IoT information group of the goods in the starting segment, and taking it as the highest benchmark outlier attribute value of the IoT information group of each goods;

[0035] The average of the highest baseline outlier attribute values ​​of the IoT information group of N goods is calculated and defined as the critical value of the IoT information outlier value of the goods.

[0036] When the outlier parameter of the IoT information of the goods in the real-time IoT information group of the goods is not lower than the critical value of the outlier of the IoT information of the goods, a message notifying that the environment in which the goods are located is bad is displayed on the client; otherwise, the message that the environment in which the goods are located is bad is not displayed.

[0037] An Internet of Things information management and control platform, comprising:

[0038] The temperature sensor, humidity sensor, pressure sensor, and client are all connected to the server for communication. The temperature sensor, humidity sensor, and pressure sensor are used to sample the temperature value, humidity value, and air pressure value of the environment where the goods are located, respectively, and transmit the sampled temperature value, humidity value, and air pressure value of the environment where the goods are located to the server. The server is used to perform processing based on the temperature value, humidity value, and air pressure value of the environment where the goods are located, and forward the temperature value, humidity value, and air pressure value of the environment where the goods are located to the client. The temperature value, humidity value, and air pressure value of the environment where the goods are located is the IoT information of the goods;

[0039] The modules running on the client include:

[0040] A cutting module is used to obtain the IoT information of each category of goods transmitted from the server and cut it into IoT information segments of multiple goods;

[0041] Identification module 1, which is used to identify the IoT information attribute vector of each IoT information segment of each commodity according to the outlier amplitude of the random change and scale change of IoT information of different commodities in the short term when outliers occur in the IoT information segment of each commodity;

[0042] The second identification module is used to identify the outlier attribute quantity of the IoT information of the goods in the IoT information group of the goods according to the IoT information attribute vector interval quantity of the goods between the IoT information segments of all the goods in the IoT information group of the goods;

[0043] The identification module three is used to identify the outlier parameters of the IoT information of the goods in the IoT information group of the goods according to the grouping arrangement of the IoT information attribute vectors of the goods in the IoT information segments of all the goods in the IoT information group of the goods and the outlier attribute quantity of the IoT information of the goods;

[0044] The detection module is used to detect the real-time Internet of Things information group of the goods based on the outlier parameters of the Internet of Things information of the goods.

[0045] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include:

[0046] According to the initial randomness and smallness of outliers generated by the IoT information of the goods, the attribute vector of the IoT information of the goods is calculated. Firstly, the fluctuation degree of the air pressure value is analyzed to improve the recognition performance of random outliers; secondly, the correlation between the temperature and humidity values ​​is used to extract the scale difference information of the temperature and humidity, so as to improve the detection sensitivity of the outlier IoT information in the environment of the goods in the future; the correlation between information of different attributes is involved; by comparing the outlier amplitude of each segment of IoT information in the IoT information group of the goods, the outlier attribute value of the IoT information of the goods is calculated, which increases the distinction between outlier information and valid IoT information and improves the recognition function of outlier information; the grouping method is used to group the goods in the IoT information group of the goods. The IoT information attribute vectors are divided into different groups, which expands the recognition space of outlier information and improves the accuracy of outlier information detection. The IoT information outlier parameters of the goods are calculated based on the number of IoT information attribute vectors of the goods in the group and the IoT information outlier attribute quantity of the goods in the IoT information group with the corresponding group centroid as the starting segment. When the IoT information outlier parameter of the goods is higher than the IoT information outlier critical quantity of the goods, a message notifying that the environment of the goods is bad is displayed on the client, which improves the application function of the IoT information of past goods and the extraction function of the IoT information attributes of goods, thereby improving the sensitivity and accuracy of IoT information outlier detection of goods. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a flow chart of the method for managing and controlling the Internet of Things information described in the present invention;

[0048] Figure 2 It is a partial structural diagram of the Internet of Things information management and control platform described in the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present invention clearer, the following will be combined with the drawings in the embodiments of the present invention to clearly and completely express the technical solution of the present invention. The embodiments expressed in the present invention 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.

[0050] like Figure 1 As shown, the Internet of Things information management and control method described in the present invention includes:

[0051] The temperature sensor, humidity sensor and pressure sensor respectively sample the temperature, humidity and air pressure values ​​of the environment in which the goods are located and transmit the sampled temperature, humidity and air pressure values ​​of the environment in which the goods are located to the server. The server performs processing based on the collected temperature, humidity and air pressure values ​​of the environment in which the goods are located and forwards the temperature, humidity and air pressure values ​​of the environment in which the goods are located to the client. The temperature, humidity and air pressure values ​​of the environment in which the goods are located are the Internet of Things information of the goods, thereby achieving the purpose of controlling the Internet of Things information of the goods; the categories of the Internet of Things information are the three categories of temperature, humidity and air pressure values ​​of the environment in which the goods are located.

[0052] The method for the client to process the IoT information of the goods includes:

[0053] S1: Obtain IoT information of each category of goods from the server and cut it into IoT information segments of multiple goods;

[0054] In a preferred but non-limiting embodiment of the present invention, the sampling rate of the temperature sensor, humidity sensor, and pressure sensor for Internet of Things information is g, and g is 400 times / second.

[0055] In a preferred but non-restrictive implementation of the present invention, in S1, the same grouping processing is performed on the IoT information of the three categories of goods obtained, and each segment formed by the IoT information of the goods is defined as the IoT information segment of the goods. The corresponding IoT information segments of the goods of each category are respectively defined as: air pressure value segment, temperature value segment and humidity value segment. Here, the number of IoT information contained in the IoT information segment of each goods is P, and P in the present invention can be one thousand.

[0056] Therefore, the Internet of Things information of each category of goods can be obtained through the above method.

[0057] S2: According to the outlier amplitude of the random change and scale change of the IoT information of different goods in the IoT information segment of each product when the outlier occurs, the IoT information attribute vector of the goods in the IoT information segment of each product is determined;

[0058] In a preferred but non-limiting embodiment of the present invention, the conditions and complexity of the environment in which the goods are located at different time points are different. When only analyzing the changes in the IoT information of the sampled goods under long-term conditions, the outliers of the IoT information of the goods cannot be efficiently extracted. In addition, the generation of the IoT information of the initial outlier goods often does not have significant timing and lasts for a long period of time. Therefore, the present invention performs analysis on the short-term change amplitude of the IoT information of the goods in each IoT information segment of the goods, calculates the IoT information attribute vector of the goods, and is used to obtain the IoT information change status of the environment in which the goods are located in a short period of time. Compared with the long-term IoT information analysis, it can more accurately identify outliers, so S2 specifically includes:

[0059] S2-1: According to the random outlier distribution of air pressure values ​​in a short period of time, determine the IoT information fluctuation attribute quantity of the IoT information segment of each item;

[0060] In a preferred but non-restrictive embodiment of the present invention, S2-1 specifically includes: when the IoT information of the goods generates an outlier, the air pressure value will fluctuate significantly due to changes in the outlier temperature and humidity. However, when the outlier occurs, the average change of the air pressure value in a long period of time is not high. On the contrary, random outliers will be generated in a period of time of the air pressure value. To improve the detection function of random outliers, the air pressure value is cleaned up with an autoencoder to remove the outliers in the air pressure value to obtain the air pressure reference IoT information.

[0061] In a preferred but non-restrictive implementation of the present invention, S2-1 specifically also includes: defining the interval between the air pressure value at the corresponding time point and the air pressure reference Internet of Things information as the air pressure interval, the more significant the difference in information value between the air pressure value and the air pressure reference Internet of Things information, the higher the air pressure interval; the interval between the air pressure value at the corresponding time point and the air pressure reference Internet of Things information can be: the sum of the moduli of the quantity obtained by subtracting the air pressure reference Internet of Things information from the air pressure value at the corresponding time point.

[0062] The present invention defines the quotient obtained by dividing the variance of the air pressure interval in the Internet of Things information segment of each commodity by the maximum amount of the air pressure interval as the Internet of Things information fluctuation attribute amount of the commodity in the Internet of Things information segment of each commodity, which is used to reflect the effect of the outlier on the Internet of Things information of the commodity.

[0063] The greater the number and amplitude of outliers generated in the IoT information segment of the goods, the more discrete the arrangement of the air pressure interval, the higher its variance, and the higher the corresponding fluctuation attribute of the IoT information of the goods. In addition, the more complex the environment in which the goods are located (for example, the strong clutter value formed by the magnetic field), the higher the amplitude of the outliers generated, and the higher the variance of the corresponding air pressure interval; while the outliers generated in a simple environment (for example, the weak clutter value formed by the magnetic field) have a low amplitude, and the variance of the corresponding air pressure interval is not high. It is necessary to synchronously detect the outlier IoT information when the environment in which the goods are located is complex and when the environment in which the goods are located is complex or not, and compare the variance of the air pressure interval in the IoT information segment of the goods with the highest amount of the air pressure interval to reflect the corresponding effect of the outliers in the IoT information segment of the goods on the IoT information of the goods.

[0064] S2-2: Determine the IoT information scale attribute quantity of the goods in the IoT information segment of each goods based on the corresponding change amplitude between the maximum scale differences of the humidity and temperature values.

[0065] In a preferred but non-restrictive embodiment of the present invention, S2-2 specifically includes: when the IoT information of the goods is outliers, the adverse effect on the goods is small, and the amplitude change of the IoT information of the goods is not high. However, the small adverse effect on the goods often significantly extends the scale difference of humidity and temperature, resulting in a reduction in the air pressure coefficient of the environment in which the goods are located, which will not only increase the adverse effect on the goods, but may even cause serious damage to the goods. Therefore, it is necessary to enhance the detection of the temperature and humidity scales of the environment in which the goods are located to improve the detection sensitivity of the outlier IoT information of the environment in which the goods are located.

[0066] The location of the temperature maximum contains the temperature scale information. In order to reduce the noise value generated by magnetic field disturbance and outlier information in the set time period during the search for the maximum value, the temperature value is processed by the autoencoder to obtain the temperature reference information. The direct method is used to find all the maximum values ​​of the temperature reference information, and the information sequence code corresponding to each maximum value is defined as the temperature maximum scale sequence code. The humidity value is processed in the same way as the temperature value to obtain the humidity reference information and the humidity maximum scale sequence code. The sequence code is the sequence value given by the information one by one according to the order of sampling.

[0067] The minimum number of maximum values ​​in the temperature reference IoT information and the humidity reference IoT information is defined as the maximum reference number s, and the modulus of the amount obtained by subtracting the corresponding temperature maximum value scaling order code from the humidity maximum value scaling order code according to the maximum reference number is defined as the scaling difference of each IoT information;

[0068] Because different goods are located in different environments with different air pressure coefficients, the corresponding change amplitudes of the temperature and humidity scale differences are used to reflect the outlier amplitude of the goods' IoT information. Therefore, the present invention defines the sum of the quotients obtained by dividing the IoT information scale differences of all goods in the IoT information segment of each goods by the IoT information scale differences of the lowest goods as the IoT information scale attribute quantity of the goods in the IoT information segment of each goods.

[0069] The more significant the extension of the scale difference of temperature and humidity is, the higher the outlier amplitude of the IoT information of the goods is, and the higher the scale attribute value of the IoT information of the goods obtained by calculation is.

[0070] Then, the IoT information fluctuation attribute quantity of the goods in the IoT information segment of each goods and the IoT information scale attribute quantity of the goods are combined to form the IoT information attribute vector of the goods in the IoT information segment of each goods, and used as the analysis object for subsequent IoT information processing.

[0071] Therefore, the IoT information attribute vector of each IoT information segment of each commodity is calculated and obtained according to the above method, which is used to reflect the outlier amplitude of the IoT information in the IoT information segment of the commodity.

[0072] S3: according to the IoT information attribute vector interval of the goods between the IoT information segments of all goods in the IoT information group of the goods, determine the IoT information outlier attribute quantity of the goods in the IoT information group of the goods;

[0073] In a preferred but non-restrictive implementation of the present invention, S3 specifically includes: when the IoT information of goods generates an outlier, the frequency of the outlier information is very small and has great randomness. It is not easy to obtain the outlier information by only analyzing the IoT information segment of a single goods. Therefore, the IoT information segment of the current goods (the IoT information segment of the current goods is the IoT information segment currently being processed) is used as the starting segment, and then the IoT information segments of N-1 goods are taken backward to form the IoT information group of the goods of the starting segment. The IoT information group of each goods contains a total of N IoT information segments of goods. In the present invention, N can be one hundred, and the information entropy between the starting segment and the IoT information segments of other goods in the IoT information group of the goods in which it is located is defined as the inter-segment spacing;

[0074] Due to the randomness and minuteness of outlier information, it is necessary to analyze the difference between the IoT information segments of reasonable goods and the IoT information segments of outlier goods. Therefore, the present invention calculates the IoT information outlier attribute quantity of the goods in the IoT information group of goods, which is:

[0075] First, the search space for outlier information needs to be expanded. In the IoT information group of the goods where the starting segment is located, the attribute vector arrangement criticality between the starting segment and the IoT information segments of other goods in the IoT information group of the goods where the starting segment is located is calculated based on the segment spacing between the IoT information segments of other goods in the IoT information group of the goods where the starting segment is located and the IoT attribute vector spacing of the goods. Here, the attribute vector arrangement criticality is inversely proportional to the segment spacing and directly proportional to the IoT attribute vector spacing of the goods.

[0076] The square root of the sum of the squares of the values ​​of the IoT information attribute vectors of the goods between the starting segment and the IoT information segments of all other goods in the IoT information group of the goods in which it is located is divided by the quotient of the corresponding inter-segment distance and is used as the attribute vector arrangement criticality between the starting segment and the IoT information segments of all other goods in the IoT information group of the goods in which it is located.

[0077] When the distance between the IoT information segment of the goods and the starting segment in the IoT information group of the goods is not low, the impact of the outlier information on the starting segment is lower, and the attribute vector arrangement criticality is lower; and when the square root of the sum of the squares of the values ​​of the IoT information attribute vector of the goods is higher, the impact on the starting segment in the group is higher, and the attribute vector arrangement criticality is higher.

[0078] Sum the attribute vectors of all other IoT information segments of the goods in the IoT information group of the goods, perform a forward aggregation such as summing the summed value and the IoT information attribute vector of the goods in the starting segment, and define the forward aggregation value as the attribute vector improvement amount of the starting segment;

[0079] In the present invention, the attribute vector arrangement criticalities of all other IoT information segments of goods in the IoT information group of goods are summed, the summed value is multiplied by the square root of the sum of the squares of the values ​​of the attribute vector of the IoT information of the goods in the starting segment to obtain an intermediate quantity, and the intermediate quantity is defined as the attribute vector improvement amount of the starting segment, which is used to reflect the relative outlier amplitude of the IoT information in the IoT information group of goods with the current IoT information segment of the goods as the starting segment.

[0080] The attribute vector improvement of the IoT information segment of each item is then calculated.

[0081] Then, multiply the improvement amount of the attribute vector of the IoT information segment of each product by the IoT information attribute vector of the IoT information segments of all other products in the IoT information group of the product, and sum them up to obtain the IoT outlier attribute amount of the product in the IoT information group of the product.

[0082] When the outlier amplitude within the IoT information segment of the goods is not small, by configuring the key degree within the IoT information group of the goods in which it is located, a larger amount of IoT information outlier attribute of the goods can be calculated, which better reflects the outlier amplitude of IoT information between different segments.

[0083] Therefore, according to the above method, the outlier attribute quantity of the IoT information of each commodity in the IoT information group is obtained, which increases the distinction between outlier information and reasonable IoT information and enhances the recognition function of outlier information.

[0084] S4: according to the grouping arrangement of the IoT information attribute vectors of all the IoT information segments of the goods in the IoT information group of the goods and the IoT information outlier attribute quantity of the goods, determine the IoT information outlier parameter of the goods in the IoT information group of the goods;

[0085] In a preferred but non-limiting embodiment of the present invention, S4 specifically includes: grouping the IoT information attribute vectors of the goods in the IoT information segments of all goods in the IoT information group of the goods using the CLARANS method to obtain multiple groups and group centroids;

[0086] The number of elements in each group in the IoT information group of a random commodity and the centroid of the group are combined as the IoT information outlier attribute quantity of the IoT information group of the commodity in the starting segment, and the IoT information outlier parameter of the IoT information group of the random commodity is determined, which is:

[0087] The number of IoT information in each group in the IoT information group of the goods is taken as the outlier attribute of the IoT information of the goods in the IoT information group of the goods at the starting segment, and the value of the forward aggregation is defined as the group outlier value of each group in the IoT information group of the goods;

[0088] In the present invention, the amount obtained by multiplying the number of IoT information in each group in the IoT information group of the goods by the centroid of the group can be used as the outlier attribute amount of the IoT information of the goods in the IoT information group of the goods at the starting segment, and can be used as the group outlier value of each group in the IoT information group of the goods.

[0089] Next, the sum of the group outlier values ​​of all groups in the IoT information group of the goods is calculated, so as to obtain the IoT information outlier parameter of the goods in the IoT information group of the goods;

[0090] First, when the outlier amplitude of the data within the group is higher, the outlier attribute of the IoT information of the goods in the centroid of each group is higher, and the outlier parameter of the IoT information of the corresponding group of goods is higher; second, when the number of IoT information segments of the goods in the corresponding group of the centroid of the group where the outlier attribute of the IoT information of the goods is not low is larger, the outlier parameter of the IoT information of the goods obtained by calculation is higher; and when the number of IoT information segments of the goods in the corresponding group of the centroid of the group where the outlier attribute of the IoT information of the goods is not high is larger, the outlier parameter of the IoT information of the goods obtained by calculation is lower.

[0091] Therefore, according to the above method, the outlier parameters of the IoT information of each commodity in the IoT information group are obtained, which expands the recognition space of outlier information and improves the accuracy of outlier information detection.

[0092] S5: Detect the real-time IoT information group of the goods based on the IoT information outlier parameters of the goods.

[0093] In a preferred but non-limiting embodiment of the present invention, S5 specifically includes: obtaining N reasonable IoT information groups of goods in the IoT information of past goods (that is, the IoT information of goods transmitted previously), and taking the centroid of all groups of the IoT information groups of each goods as the highest amount in the IoT information outlier attribute amount of the goods in the IoT information group of the goods in the starting segment, and taking it as the highest benchmark outlier attribute amount of the IoT information group of each goods;

[0094] The average of the highest baseline outlier attribute values ​​of the IoT information group of N goods is calculated and defined as the critical value of the IoT information outlier value of the goods.

[0095] When the outlier parameter of the IoT information of the goods in the real-time IoT information group of the goods is not lower than the critical value of the outlier of the IoT information of the goods, a message notifying that the environment in which the goods are located is bad is displayed on the client; otherwise, the message that the environment in which the goods are located is bad is not displayed.

[0096] like Figure 2 As shown, the Internet of Things information management and control platform described in the present invention includes:

[0097] The temperature sensor, humidity sensor, pressure sensor, and client are all connected to the server for communication. The temperature sensor, humidity sensor, and pressure sensor are used to sample the temperature value, humidity value, and air pressure value of the environment in which the goods are located, respectively, and transmit the sampled temperature value, humidity value, and air pressure value of the environment in which the goods are located to the server. The server is used to perform processing based on the temperature value, humidity value, and air pressure value of the environment in which the goods are located, and forward the temperature value, humidity value, and air pressure value of the environment in which the goods are located to the client. The temperature value, humidity value, and air pressure value of the environment in which the goods are located is the IoT information of the goods, thereby achieving the purpose of controlling the IoT information of the goods;

[0098] The modules running on the client include:

[0099] A cutting module is used to obtain the IoT information of each category of goods transmitted from the server and cut it into IoT information segments of multiple goods;

[0100] Identification module 1, which is used to identify the IoT information attribute vector of each IoT information segment of each commodity according to the outlier amplitude of the random change and scale change of IoT information of different commodities in the short term when outliers occur in the IoT information segment of each commodity;

[0101] The second identification module is used to identify the outlier attribute quantity of the IoT information of the goods in the IoT information group of the goods according to the IoT information attribute vector interval quantity of the goods between the IoT information segments of all the goods in the IoT information group of the goods;

[0102] The identification module three is used to identify the outlier parameters of the IoT information of the goods in the IoT information group of the goods according to the grouping arrangement of the IoT information attribute vectors of the goods in the IoT information segments of all the goods in the IoT information group of the goods and the outlier attribute quantity of the IoT information of the goods;

[0103] The detection module is used to detect the IoT information group of the goods in real time according to the IoT information outlier parameter of the goods. The client can be a computer.

[0104] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include:

[0105] According to the initial randomness and smallness of outliers generated by the IoT information of the goods, the attribute vector of the IoT information of the goods is calculated. Firstly, the fluctuation degree of the air pressure value is analyzed to improve the recognition performance of random outliers; secondly, the correlation between the temperature and humidity values ​​is used to extract the scale difference information of the temperature and humidity, so as to improve the detection sensitivity of the outlier IoT information in the environment of the goods in the future; the correlation between information of different attributes is involved; by comparing the outlier amplitude of each segment of IoT information in the IoT information group of the goods, the outlier attribute value of the IoT information of the goods is calculated, which increases the distinction between outlier information and valid IoT information and improves the recognition function of outlier information; the grouping method is used to group the goods in the IoT information group of the goods. The IoT information attribute vectors are divided into different groups, which expands the recognition space of outlier information and improves the accuracy of outlier information detection. The IoT information outlier parameters of the goods are calculated based on the number of IoT information attribute vectors of the goods in the group and the IoT information outlier attribute quantity of the goods in the IoT information group with the corresponding group centroid as the starting segment. When the IoT information outlier parameter of the goods is higher than the IoT information outlier critical quantity of the goods, a message notifying that the environment of the goods is bad is displayed on the client, which improves the application function of the IoT information of past goods and the extraction function of the IoT information attributes of goods, thereby improving the sensitivity and accuracy of IoT information outlier detection of goods.

[0106] 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 included in the protection scope of the claims of the present invention.

Claims

1. A method for managing and controlling information of the Internet of Things, characterized in that: include: The temperature sensor, humidity sensor, and pressure sensor respectively sample the temperature value, humidity value, and air pressure value of the environment where the goods are located, and transmit the sampled temperature value, humidity value, and air pressure value of the environment where the goods are located to the server. The server performs processing based on the temperature value, humidity value, and air pressure value of the environment where the goods are located, and forwards the temperature value, humidity value, and air pressure value of the environment where the goods are located to the client. The temperature value, humidity value, and air pressure value of the environment where the goods are located is the IoT information of the goods; The method for the client to process the IoT information of the goods includes: S1: Obtain IoT information of each category of goods from the server and cut it into IoT information segments of multiple goods; S2: According to the outlier amplitude of the random change and scale change of the IoT information of different goods in the IoT information segment of each product when the outlier occurs, the IoT information attribute vector of the goods in the IoT information segment of each product is determined; S3: according to the IoT information attribute vector interval of the goods between the IoT information segments of all goods in the IoT information group of the goods, determine the IoT information outlier attribute quantity of the goods in the IoT information group of the goods; S4: according to the grouping arrangement of the IoT information attribute vectors of all the IoT information segments of the goods in the IoT information group of the goods and the IoT information outlier attribute quantity of the goods, determine the IoT information outlier parameter of the goods in the IoT information group of the goods; S5: Detect the real-time IoT information group of the goods based on the IoT information outlier parameters of the goods.

2. The method for managing and controlling Internet of Things information according to claim 1, characterized in that: The sampling rate of the temperature sensor, humidity sensor, and pressure sensor for IoT information is g, which is 400 times / second; In S1, the same grouping processing is performed on the IoT information of the three categories of goods obtained, and each segment formed by the IoT information of the goods is defined as the IoT information segment of the goods. The corresponding IoT information segments of the goods of each category are respectively defined as: air pressure value segment, temperature value segment and humidity value segment. Here, the number of IoT information contained in each IoT information segment of the goods is P.

3. The method for managing and controlling Internet of Things information according to claim 2, characterized in that: S2 specifically includes: S2-1: According to the random outlier distribution of air pressure values ​​in a short period of time, determine the IoT information fluctuation attribute quantity of the IoT information segment of each item; S2-2: Determine the IoT information scale attribute quantity of the goods in the IoT information segment of each goods based on the corresponding change amplitude between the maximum scale differences of the humidity and temperature values.

4. The method for managing and controlling Internet of Things information according to claim 3, characterized in that: S2-1 specifically includes: using an autoencoder to remove outliers in the air pressure value to obtain air pressure benchmark IoT information.

5. The method for managing and controlling Internet of Things information according to claim 4, characterized in that: S2-1 specifically includes: defining the interval between the air pressure value at the corresponding time point and the air pressure reference IoT information as the air pressure interval; The quotient obtained by dividing the variance of the air pressure interval in the Internet of Things information segment of each commodity by the maximum amount of the air pressure interval is defined as the Internet of Things information fluctuation attribute amount of the commodity in the Internet of Things information segment of each commodity.

6. The method for managing and controlling Internet of Things information according to claim 5, characterized in that: S2-2 specifically includes: using the autoencoder to process the temperature value to obtain the temperature reference IoT information, using the direct method to find out all the maximum values ​​of the temperature reference IoT information, and defining the IoT information sequence code corresponding to each maximum value as the temperature maximum value scaling sequence code, and using the same processing method as the temperature value to process the humidity value to obtain the humidity reference IoT information and the humidity maximum value scaling sequence code; The minimum number of maximum values ​​in the temperature reference IoT information and the humidity reference IoT information is defined as the maximum reference number s, and the modulus of the amount obtained by subtracting the corresponding temperature maximum value scaling order code from the humidity maximum value scaling order code according to the maximum reference number is defined as the scaling difference of each IoT information; The sum of the quotients obtained by dividing the IoT information scaling differences of all goods in the IoT information segment of each goods by the IoT information scaling difference of the lowest goods is defined as the IoT information scaling attribute quantity of the goods in the IoT information segment of each goods.

7. The method for managing and controlling Internet of Things information according to claim 6, characterized in that: S3 specifically includes: using the IoT information segment of the current goods as the starting segment, then taking the IoT information segments of N-1 goods backward to form the IoT information group of the goods in the starting segment, each IoT information group of the goods contains a total of N IoT information segments of the goods, and defining the information entropy between the starting segment and the IoT information segments of other goods in the IoT information group of the goods in which it is located as the inter-segment spacing; The outlier attribute amount of the IoT information of the goods in the IoT information group of the goods is calculated, which is: In the IoT information group of the goods where the starting segment is located, the attribute vector arrangement criticality between the starting segment and each other IoT information segment of the goods in the IoT information group where the starting segment is located is calculated according to the segment spacing between the starting segment and each other IoT information segment of the goods in the IoT information group where the starting segment is located and the IoT information attribute vector spacing of the goods; Sum the attribute vector arrangement criticalities of all other goods' IoT information segments in the goods' IoT information group, perform forward aggregation on the sum value and the IoT information attribute vector of the starting segment, and define the forward aggregation value as the attribute vector improvement amount of the starting segment.

8. The method for managing and controlling Internet of Things information according to claim 7, characterized in that: S4 specifically includes: grouping the IoT information attribute vectors of all IoT information segments of the goods in the IoT information group of the goods using the CLARANS method to obtain multiple groups and group centroids; The number of elements in each group in the IoT information group of a random commodity and the centroid of the group are combined as the IoT information outlier attribute quantity of the IoT information group of the commodity in the starting segment, and the IoT information outlier parameter of the IoT information group of the random commodity is determined, which is: The number of IoT information in each group in the IoT information group of the goods is taken as the outlier attribute of the IoT information of the goods in the IoT information group of the goods at the starting segment, and the value of the forward aggregation is defined as the group outlier value of each group in the IoT information group of the goods; Next, the sum of the group outlier values ​​of all groups in the IoT information group of the goods is calculated to obtain the IoT information outlier parameters of the goods in the IoT information group of the goods.

9. The method for managing and controlling Internet of Things information according to claim 8, characterized in that: S5 specifically includes: obtaining N reasonable IoT information groups of goods in the IoT information of past goods, taking the centroid of all groups of the IoT information groups of each goods as the highest value in the IoT information outlier attribute value of the goods in the IoT information group of the goods in the starting segment, and taking it as the highest benchmark outlier attribute value of the IoT information group of each goods; The average of the highest baseline outlier attribute values ​​of the IoT information group of N goods is calculated and defined as the critical value of the IoT information outlier value of the goods. When the outlier parameter of the IoT information of the goods in the real-time IoT information group of the goods is not lower than the critical value of the outlier of the IoT information of the goods, a message notifying that the environment in which the goods are located is bad is displayed on the client; otherwise, the message that the environment in which the goods are located is bad is not displayed.

10. An Internet of Things information management and control platform, characterized in that: include: The temperature sensor, humidity sensor, pressure sensor, and client are all connected to the server for communication. The temperature sensor, humidity sensor, and pressure sensor are used to sample the temperature value, humidity value, and air pressure value of the environment where the goods are located, respectively, and transmit the sampled temperature value, humidity value, and air pressure value of the environment where the goods are located to the server. The server is used to perform processing based on the temperature value, humidity value, and air pressure value of the environment where the goods are located, and forward the temperature value, humidity value, and air pressure value of the environment where the goods are located to the client. The temperature value, humidity value, and air pressure value of the environment where the goods are located is the IoT information of the goods; The modules running on the client include: A cutting module is used to obtain the IoT information of each category of goods transmitted from the server and cut it into IoT information segments of multiple goods; Identification module 1, which is used to identify the IoT information attribute vector of each IoT information segment of each commodity according to the outlier amplitude of the random change and scale change of IoT information of different commodities in the short term when outliers occur in the IoT information segment of each commodity; The second identification module is used to identify the outlier attribute quantity of the IoT information of the goods in the IoT information group of the goods according to the IoT information attribute vector interval quantity of the goods between the IoT information segments of all the goods in the IoT information group of the goods; The identification module three is used to identify the outlier parameters of the IoT information of the goods in the IoT information group of the goods according to the grouping arrangement of the IoT information attribute vectors of the goods in the IoT information segments of all the goods in the IoT information group of the goods and the outlier attribute quantity of the IoT information of the goods; The detection module is used to detect the real-time Internet of Things information group of the goods based on the outlier parameters of the Internet of Things information of the goods.

Images