An equipment control management system and method based on AI artificial intelligence

Abstract

The present invention discloses an equipment control management system and method based on AI artificial intelligence, belonging to the field of intelligent control technology. The present invention collects different data collected by different sensors when different instructions are generated in the history of intelligent devices, and calculates the decision ratio of different data to the instructions generated by intelligent devices; compares and analyzes different instructions generated in the history of intelligent devices, calculates the data decision ratio threshold when the instructions change and the standard time threshold when the device generates instructions, and extracts the features of historical instructions, monitors the transmission time of data collected by different sensors to the central processor in real time, judges whether the data is delayed according to the standard time threshold, and obtains delayed data; judges whether the delayed data changes the instructions generated by the intelligent device according to the decision ratio threshold; then calculates the similarity between the real-time data and each instruction, generates the corresponding output instruction after comparison, and warns the delayed data.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and specifically to an equipment control management system and method based on AI artificial intelligence. Background Art

[0002] In the modern society where science and technology are developing rapidly, the use of smart devices in life is becoming more and more frequent. The development of smart devices began with early computers. In the mid-20th century, the first general-purpose computers appeared, which were mainly used to solve complex mathematical and scientific computing problems. The intelligence level of these computers was relatively low, but they laid the foundation for later smart devices. In the 1970s and 1980s, expert systems became a key research direction in the field of artificial intelligence. These systems are based on professional knowledge and are designed to solve problems in specific fields. In the 1990s, the widespread use of the Internet promoted the rapid development of information technology. Smart devices began to connect to the Internet, and people could access and control various devices through the network. With the advancement of mobile technology, smartphones and tablets have become an indispensable part of people's daily lives. These devices integrate a variety of sensors and advanced processing capabilities to support a variety of applications and services. Various smart devices collect various data through various sensors. Smart devices integrate and analyze various data to generate different instructions and complete different tasks and movements. However, after an abnormal situation occurs, the smart device cannot receive the data that needs to be analyzed in time, and cannot accurately generate different instructions to complete the task. In addition, since different data have different impacts on the generation of instructions by smart devices, the delay of some data will not affect the work of smart devices, and smart devices cannot determine which data will change the instructions and which data will not change the instructions. Some smart devices will stop working after they cannot receive data, which greatly reduces work efficiency, and some smart devices will not stop working, but may generate wrong instructions to cause greater losses. Therefore, the present invention provides an equipment control management system and method based on AI artificial intelligence to solve the above problems. Summary of the invention

[0003] The purpose of the present invention is to provide an equipment control management system and method based on AI artificial intelligence to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0005] An AI-based equipment control and management method, the method comprising the following steps:

[0006] S100, collecting different data collected by different sensors when different instructions are generated in the history of the smart device, analyzing the different data of each instruction collected from the different instructions, and obtaining a decision ratio of each data when the smart device generates different instructions;

[0007] Furthermore, the specific steps for obtaining the decision ratio of different data to generate instructions for smart devices are as follows:

[0008] S101, marking the data corresponding to different instructions generated in the history of the smart device, and generating a data set S as {s 1 、s 2 、s 3 ...s n},s 1 、s 2 、s 3 ...s n are the 1st, 2nd, 3rd...nth types of data in different instructions, and n is a positive integer; suppose the total number of different instructions collected is k, and the data in the k instructions are collected and marked, and finally k different data sets are generated, and each data in each data set is different;

[0009] S102, after collecting the data corresponding to different instructions in the smart device, compare all the data sets corresponding to the different instructions, calculate the number of times p of each data change when the instructions are different, sort each data from small to large according to the calculated number of changes, and obtain the number of changes of each data after sorting as {p 1 、p 2 、p 3 …p n}, p 1 、p 2 、p 3 ...p n is the number of changes of the data in the 1st, 2nd, 3rd...nth order after sorting, where n is a positive integer;

[0010] S103, calculating the decision ratio of each data according to the number of changes of each data in different instructions, the formula is:

[0011]

[0012] E in the formula i is the decision ratio of the data in the ith order, p i is the number of changes of the i-th data in all different instructions, and k is the number of all different instructions generated historically.

[0013] The above-mentioned calculated data decision ratio is the influence ratio of each data on the smart device when generating instructions. Because the smart device generates instructions by analyzing the data collected by various sensors to obtain the corresponding instructions, the influence ratio of each data on the instructions is crucial data;

[0014] S200, comparing and analyzing different instructions generated in the history of the smart device, calculating the total decision ratio of all data that have changed when the instruction changes according to the decision ratios of different data, and finally obtaining the data decision ratio threshold when the instruction changes;

[0015] Furthermore, the specific steps for finally obtaining the data decision ratio threshold when the instruction changes are as follows:

[0016] S201, collect k different instructions generated by the smart device in history, and obtain different instruction sets as {x 1 、x 2 、x 3 ...x k}, x 1 、x 2 、x 3 ...x k The instructions generated for the 1st, 2nd, 3rd, ...kth smart devices collected, where k is a positive integer;

[0017] S202, compare the data in each instruction, when the instruction changes, extract the changed data set in the changed instruction as {s_b 1 、s_b 2 、s_b 3 ...s_b z},s_b 1 、s_b 2 、s_b 3 ...s_b z is the 1st, 2nd, 3rd...zth data that changes when the instruction changes, and z is a positive integer;

[0018] S203, obtain the changed data set {s_b 1 、s_b 2 、s_b 3 ...s_b z} and the decision ratio of each data to calculate the total decision ratio of all changed data, the formula is:

[0019]

[0020] In the formula, E_a is the total decision ratio of all data that change when the instruction changes, and E r is the decision ratio of each type of data that changes, r takes values ​​from 1 to z;

[0021] S204. Calculate the total data decision ratio in k instructions, then calculate the average value, and set the average value as the data decision ratio threshold E_y when the instruction changes.

[0022] The decision ratio threshold calculated above indicates that when the data in a certain instruction changes and the total decision ratio of the changed data is greater than this threshold, the smart device will generate different instructions. Calculating the decision ratio threshold can determine the changes in the instructions of the smart device in real time.

[0023] There are different instructions in the collected historical instructions generated by smart devices, and when each different instruction is generated, the collected data values ​​and data types are different, and the data when generating one instruction is also divided into multiple types. Therefore, the decision ratio of multiple data in different instructions can indicate the degree of influence of each data on each instruction.

[0024] S300, after calculating the data decision ratio threshold when the instruction changes, calculate the standard time threshold when the smart device generates the instruction according to the time records of different data transmissions when the smart device generates the instruction in history;

[0025] Furthermore, the specific steps for calculating the standard time threshold for data transmission required for the smart device to generate an instruction are as follows:

[0026] S301, collecting the time required for the smart device to generate instructions k times in history, specifically {t 1 ,t 2 ,t 3 ...t k}, t 1 ,t 2 ,t 3 ...t k The time required to generate the instruction for the 1st, 2nd, 3rd, ...kth time, where k is a positive integer;

[0027] S302 , calculating the average time value of the time required for the smart device to generate instructions for the k times collected, and setting the average time value as the standard time threshold t_y for the smart device to generate instructions.

[0028] When smart devices generate instructions based on various data, if some data transmission time is too long or some data cannot be obtained, the instructions generated by the smart devices will be erroneous. Therefore, the calculation of the standard time for data transmission can determine in real time whether the data is abnormal or delayed.

[0029] S400, analyzing all instructions generated in the history of the smart device to obtain different features of different instructions, and generating an instruction feature set including all instruction features;

[0030] Furthermore, the specific steps for obtaining different features of different instructions are:

[0031] S401, collecting all instructions generated by the smart device in history and classifying the instructions, classifying the same instructions into the same category, and obtaining the different instruction sets in S201;

[0032] S402, using the change data set {s_b 1 、s_b 2 、s_b 3 ...s_b z}, sort the changed data from small to large according to the decision ratio of each data in the changed data set, and set the decision ratio of the changed data in each instruction change to be {E_s_b 1 、E_s_b 2 、E_s_b 3 ...E_s_b z}, E_s_b 1 、E_s_b 2 、E_s_b 3 ...E_s_b z is the decision ratio of the changed data in the 1st, 2nd, 3rd...zth order in each instruction change, where z is a positive integer; the data type with the largest decision ratio among the changed data sets in the changed instruction is selected as the feature E_s_b of the corresponding instruction z ;

[0033] S403, calculate the features of k different instructions, and generate an instruction feature set {E_s_b z1 、E_s_b z2 、E_s_b z3 ...E_s_b zk},E_s_b z1 、E_s_b z2 、E_s_b z3 ...E_s_b zk are the characteristics of the 1st, 2nd, 3rd...kth instructions, where k is a positive integer.

[0034] S500, there is a central processor in the smart device. When the smart device is working, the central processor generates different instructions after receiving different data, monitors the transmission time of data collected by different sensors to the central processor in real time, determines whether the data is delayed according to the standard time threshold, and obtains delayed data;

[0035] Furthermore, the specific steps to obtain the delay data are:

[0036] S501, real-time acquisition of the transmission time of different data collected by different sensors, assuming that the collected data transmission time is {t_s 1 、t_s 2 、t_s 3 ...t_s n}, t_s 1 t_s 2 、t_s 3 ...t_s n is the transmission time of the 1st, 2nd, 3rd...nth data collected in real time, where n is a positive integer;

[0037] S502, determine when t_s i > t_y, there is a delay in data transmission time, let the delayed data s_d be the data s corresponding to the transmission time i , the value of i is from 1 to n.

[0038] S600, when it is determined that the data is delayed, determining whether the delayed data causes the instruction generated by the smart device to change according to the decision ratio threshold;

[0039] Furthermore, the specific steps of determining whether the delay data changes the instruction generated by the smart device by using the threshold are as follows:

[0040] S601, when real-time monitoring is performed on instructions generated by the smart device, it is determined that there is delayed data s_d;

[0041] S602. The decision ratio of the extracted delay data is E_d. When it is judged that E_d≤E_y, the delay data will not change the instructions generated by the smart device; when it is judged that E_d>E_y, the delay data will change the instructions generated by the smart device, and then the smart device stops generating instructions and issues an early warning to prompt the staff that the smart device is abnormal.

[0042] S700. When it is determined that the delayed data will not change the instructions generated by the smart device, the similarity between the real-time collected data and each different instruction in the history is calculated based on the instruction feature set, and the instruction with the greatest similarity is selected as the output instruction through comparison, and an early warning is issued for the delayed data.

[0043] Furthermore, the specific steps for calculating the similarity between the real-time collected data and each different instruction in the history are:

[0044] S701. After determining that the delayed data will not change the instruction generated by the smart device, sort the real-time collected data, exclude the calculated delayed data, and use the remaining data to calculate the similarity between the collected data and each instruction. Specifically, the remaining data is {s_u 1 、s_u 2 、s_u 3 ...s_uj}, s_u 1 、s_u 2 、s_u 3 ...s_u j The first, second, third, ...jth data collected in real time after excluding delayed data, where j is a positive integer;

[0045] S702, compare the real-time collected data set with the data corresponding to the instructions generated by the smart device in the history, calculate the number of data with the same data type as f, then compare the value of each data, and set the number of data with the same data value as w;

[0046] S703, extracting the decision ratios corresponding to the real-time collected data, sorting the real-time collected data in ascending order of the decision ratios, and obtaining the data with the largest decision ratio among the real-time collected data, whose decision ratio is E_max;

[0047] S704, calculate the similarity between the real-time collected data and the historical data corresponding to k types of instructions generated by the intelligent device, and the calculation formula is:

[0048]

[0049] In the formula, ρ is the similarity between the real-time collected data and each instruction generated by the historical intelligent device, n is the number of data corresponding to each instruction generated by the intelligent device in the history, and E h is the data decision ratio corresponding to each instruction generated by the intelligent device in history, h ranges from 1 to n, f is the number of data with the same data type compared with the data corresponding to the instructions generated by the historical intelligent device in the real-time collected data, and w is the number of data with the same data value compared with the data corresponding to the instructions generated by the historical intelligent device in the real-time collected data; E_s_b z Characterize each instruction generated by the intelligent device in history;

[0050] S705: Calculate the k instructions generated by the smart device in history in sequence according to the above formula to obtain k similarities {ρ 1 , 2 , 3 ... k} 1 , 2 , 3 ... k is the calculated similarity of the 1st, 2nd, 3rd, ...kth data, where k is a positive integer;

[0051] S706. After calculating the similarity set, the similarities in the set are compared, and the instruction corresponding to the maximum similarity is selected as the instruction Q generated after the delayed data exists. The intelligent device outputs the instruction Q and issues an early warning for the delayed data.

[0052] If data anomalies or delays occur in smart devices, the smart devices may not be able to generate instructions normally, or may generate erroneous instructions. Therefore, the above method first determines whether the delayed data changes the data, and then generates instructions through similarity. This allows the smart device to continue to generate instructions and work after an anomaly occurs. If the data is less affected, it only needs to issue an early warning for the delayed data, which will not affect the work of the smart device. When it is determined that the delayed data will change the instructions generated by the smart device, the smart device will stop working immediately and issue an early warning to avoid greater losses caused by erroneous instructions generated by the smart device.

[0053] An equipment control and management system based on AI artificial intelligence, the equipment control and management system includes a data collection module, a threshold calculation module, a feature extraction module, a judgment and early warning module, a similarity calculation module and an instruction output module;

[0054] The data collection module is used to collect historical instructions and corresponding data generated by smart devices, and to collect data detected by sensors in smart devices in real time;

[0055] The threshold calculation module is used to calculate the decision ratio threshold and the standard time threshold according to the instructions and data generated by the intelligent device in history;

[0056] The feature extraction module is used to calculate the data decision ratio corresponding to the instructions generated by the intelligent device in the collected history, and obtain the features of different instructions;

[0057] The judgment and early warning module is used to judge the real-time collected data according to the calculated decision ratio threshold and standard time threshold;

[0058] The similarity calculation module is used to calculate the similarity between the real-time collected data and each instruction after determining the delay data, and generate the corresponding instruction according to the similarity;

[0059] The instruction output module is used to generate instructions through calculation, and then have the intelligent device output the generated instructions and issue an early warning for delayed data.

[0060] The threshold calculation module includes a standard time threshold calculation unit and a decision ratio threshold calculation unit;

[0061] The standard time threshold calculation unit is used to collect the time required for the smart device to generate instructions k times in history, calculate the average time value of the collected k times required for the smart device to generate instructions, and set the average time value as the standard time threshold for the smart device to generate instructions;

[0062] The decision ratio threshold calculation unit is used to collect k different instructions generated by the intelligent device in history, compare the data in each instruction, and when the instruction changes, extract the changed data set in the changed instruction, analyze the data set, calculate the total decision ratio of the data in the k instructions, and then calculate the average value, and set the average value as the data decision ratio threshold when the instruction changes;

[0063] The judgment and early warning module includes a delay data judgment unit and a decision ratio judgment unit;

[0064] The delayed data judgment unit is used to collect the transmission time of different data collected by different sensors in real time, and compare the transmission time of real-time data with the standard time threshold to determine whether there is delayed data;

[0065] The decision ratio judgment unit is used to monitor the instructions generated by the smart device in real time, and after determining that there is delay data, extract the decision ratio of the delay data, compare the decision ratio of the delay data with the decision ratio threshold, and determine whether the delay data will change the instructions generated by the smart device. When it is determined that the delay data will change the instructions generated by the smart device, the instruction feature similarity is calculated based on each data collected in real time, and the corresponding instructions are generated; when it is determined that the delay data will not change the instructions generated by the smart device, the smart device stops generating instructions and issues an early warning, prompting the staff of the smart device abnormality.

[0066] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0067] 1. The present invention calculates the decision ratio of the corresponding data when the instruction is generated in the smart device, so as to clearly know the influence of each data on the instruction generated by the smart device, and to a certain extent avoid the smart device from generating erroneous instructions and reduce losses.

[0068] 2. The present invention can determine whether data anomalies occur in the smart device and whether the abnormal data will cause the smart device to generate erroneous instructions by judging the transmission time and decision ratio of each type of data. It can also determine whether the abnormal data will cause the smart device to generate erroneous instructions. It can timely discover the abnormalities of the smart device and avoid losses caused by abnormalities in the smart device during operation.

[0069] 3. The present invention calculates the similarity between real-time collected data and each instruction generated by the historical intelligent device, so that when data anomalies occur in the intelligent device, but the abnormal data cannot change the instructions, the correct instructions can be accurately generated, so that the intelligent device can continue to work. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0071] Figure 1 This is a flowchart of the steps of an equipment control management method based on AI artificial intelligence of the present invention;

[0072] Figure 2 It is a module schematic diagram of an equipment control management system based on AI artificial intelligence of the present invention. DETAILED DESCRIPTION

[0073] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0074] See also Figure 1-Figure 2 , the present invention provides a technical solution:

[0075] An AI-based equipment control and management method, the method comprising the following steps:

[0076] S100, collecting different data collected by different sensors when different instructions are generated in the history of the smart device, analyzing the different data of each instruction collected from the different instructions, and obtaining a decision ratio of each data when the smart device generates different instructions;

[0077] Furthermore, the specific steps for obtaining the decision ratio of different data to generate instructions for smart devices are as follows:

[0078] S101, marking the data corresponding to different instructions generated in the history of the smart device, and generating a data set S as {s 1 、s 2 、s 3 ...s n},s 1 、s 2 、s 3 ...s n are the 1st, 2nd, 3rd...nth types of data in different instructions, and n is a positive integer; suppose the total number of different instructions collected is k, and the data in the k instructions are collected and marked, and finally k different data sets are generated, and each data in each data set is different;

[0079] S102, after collecting the data corresponding to different instructions in the smart device, compare all the data sets corresponding to the different instructions, calculate the number of times p of each data change when the instructions are different, sort each data from small to large according to the calculated number of changes, and obtain the number of changes of each data after sorting as {p1 、p 2 、p 3 ...p n}, p 1 、p 2 、p 3 ...p n is the number of changes of the data in the 1st, 2nd, 3rd...nth order after sorting, where n is a positive integer;

[0080] S103, calculating the decision ratio of each data according to the number of changes of each data in different instructions, the formula is:

[0081]

[0082] E in the formula i is the decision ratio of the data in the ith order, p i is the number of changes of the i-th data in all different instructions, and k is the number of all different instructions generated historically.

[0083] The above-mentioned calculated data decision ratio is the influence ratio of each data on the smart device when generating instructions. Because the smart device generates instructions by analyzing the data collected by various sensors to obtain the corresponding instructions, the influence ratio of each data on the instructions is crucial data;

[0084] There are different instructions in the collected historical instructions generated by smart devices, and when each different instruction is generated, the collected data values ​​and data types are different, and the data when generating one instruction is also divided into multiple types. Therefore, the decision ratio of multiple data in different instructions can indicate the degree of influence of each data on each instruction.

[0085] S200, comparing and analyzing different instructions generated in the history of the smart device, calculating the total decision ratio of all data that have changed when the instruction changes according to the decision ratios of different data, and finally obtaining the data decision ratio threshold when the instruction changes;

[0086] Furthermore, the specific steps for finally obtaining the data decision ratio threshold when the instruction changes are as follows:

[0087] S201, collect k different instructions generated by the smart device in history, and obtain different instruction sets as {x 1 、x 2 、x 3 ...x k}, x 1 、x 2 、x 3 ...x k The instructions generated for the 1st, 2nd, 3rd, ...kth smart devices collected, where k is a positive integer;

[0088] S202, compare the data in each instruction, when the instruction changes, extract the changed data set in the changed instruction as {s_b 1 、s_b 2 、s_b 3 ...s_b z},s_b 1 、s_b 2 、s_b 3 ...s_b z is the 1st, 2nd, 3rd...zth data that changes when the instruction changes, and z is a positive integer;

[0089] S203, obtain the changed data set {s_b 1 、s_b 2 、s_b 3 ...s_b z} and the decision ratio of each data to calculate the total decision ratio of all changed data, the formula is:

[0090]

[0091] In the formula, E_a is the total decision ratio of all data that change when the instruction changes, and E r is the decision ratio of each type of data that changes, r takes values ​​from 1 to z;

[0092] S204. Calculate the total data decision ratio in k instructions, then calculate the average value, and set the average value as the data decision ratio threshold E_y when the instruction changes.

[0093] The decision ratio threshold calculated above indicates that when the data in a certain instruction changes and the total decision ratio of the changed data is greater than this threshold, the smart device will generate different instructions. Calculating the decision ratio threshold can determine the changes in the instructions of the smart device in real time.

[0094] S300, after calculating the data decision ratio threshold when the instruction changes, calculate the standard time threshold when the smart device generates the instruction according to the time records of different data transmissions when the smart device generates the instruction in history;

[0095] Furthermore, the specific steps for calculating the standard time threshold for data transmission required for the smart device to generate an instruction are as follows:

[0096] S301, collecting the time required for the smart device to generate instructions k times in history, specifically {t 1 ,t 2 ,t 3 ...t k}, t 1 ,t 2 ,t3 ...t k The time required to generate the instruction for the 1st, 2nd, 3rd, ...kth time, where k is a positive integer;

[0097] S302 , calculating the average time value of the time required for the smart device to generate instructions for the k times collected, and setting the average time value as the standard time threshold t_y for the smart device to generate instructions.

[0098] When smart devices generate instructions based on various data, if some data transmission time is too long or some data cannot be obtained, the instructions generated by the smart devices will be erroneous. Therefore, the calculation of the standard time for data transmission can determine in real time whether the data is abnormal or delayed.

[0099] S400, analyzing all instructions generated in the history of the smart device to obtain different features of different instructions, and generating an instruction feature set including all instruction features;

[0100] Furthermore, the specific steps for obtaining different features of different instructions are:

[0101] S401, collecting all instructions generated by the smart device in history and classifying the instructions, classifying the same instructions into the same category, and obtaining the different instruction sets in S201;

[0102] S402, using the change data set {s_b 1 、s_b 2 、s_b 3 ...s_b z}, sort the changed data from small to large according to the decision ratio of each data in the changed data set, and set the decision ratio of the changed data in each instruction change to be {E_s_b 1 、E_s_b 2 、E_s_b 3 ...E_s_b z}, E_s_b 1 、E_s_b 2 、E_s_b 3 ...E_s_b z is the decision ratio of the changed data in the 1st, 2nd, 3rd...zth order in each instruction change, where z is a positive integer; the data type with the largest decision ratio among the changed data sets in the changed instruction is selected as the feature E_s_b of the corresponding instruction z ;

[0103] S403, calculate the features of k different instructions, and generate an instruction feature set {E_s_b z1 、E_s_b z2、E_s_b z3 ...E_s_b zk},E_s_b z1 、E_s_b z2 、E_s_b z3 ...E_s_b zk are the characteristics of the 1st, 2nd, 3rd...kth instructions, where k is a positive integer.

[0104] S500, there is a central processor in the smart device. When the smart device is working, the central processor generates different instructions after receiving different data, monitors the transmission time of data collected by different sensors to the central processor in real time, determines whether the data is delayed according to the standard time threshold, and obtains delayed data;

[0105] Furthermore, the specific steps to obtain the delay data are:

[0106] S501, real-time acquisition of the transmission time of different data collected by different sensors, assuming that the collected data transmission time is {t_s 1 、t_s 2 、t_s 3 ...t_s n}, t_s 1 、t_s 2 、t_s 3 ...t_s n is the transmission time of the 1st, 2nd, 3rd...nth data collected in real time, where n is a positive integer;

[0107] S502, determine when t_s i > t_y, there is a delay in data transmission time, let the delayed data s_d be the data s corresponding to the transmission time i , the value of i is from 1 to n.

[0108] S600, when it is determined that the data is delayed, determining whether the delayed data causes the instruction generated by the smart device to change according to the decision ratio threshold;

[0109] Furthermore, the specific steps of determining whether the delay data changes the instruction generated by the smart device by using the threshold are as follows:

[0110] S601, when real-time monitoring is performed on instructions generated by the smart device, it is determined that there is delayed data s_d;

[0111] S602. The decision ratio of the extracted delay data is E_d. When it is judged that E_d≤E_y, the delay data will not change the instructions generated by the smart device; when it is judged that E_d>E_y, the delay data will change the instructions generated by the smart device, and then the smart device stops generating instructions and issues an early warning to prompt the staff that the smart device is abnormal.

[0112] S700. When it is determined that the delayed data will not change the instructions generated by the smart device, the similarity between the real-time collected data and each different instruction in the history is calculated based on the instruction feature set, and the instruction with the greatest similarity is selected as the output instruction through comparison, and an early warning is issued for the delayed data.

[0113] Furthermore, the specific steps for calculating the similarity between the real-time collected data and each different instruction in the history are:

[0114] S701. After determining that the delayed data will not change the instruction generated by the smart device, sort the real-time collected data, exclude the calculated delayed data, and use the remaining data to calculate the similarity between the collected data and each instruction. Specifically, the remaining data is {s_u 1 、s_u 2 、s_u 3 ...s_u j},s_u 1 、s_u 2 、s_u3...s_u j The first, second, third, ...jth data collected in real time after excluding delayed data, where j is a positive integer;

[0115] S702, compare the real-time collected data set with the data corresponding to the instructions generated by the smart device in the history, calculate the number of data with the same data type as f, then compare the value of each data, and set the number of data with the same data value as w;

[0116] S703, extracting the decision ratios corresponding to the real-time collected data, sorting the real-time collected data in ascending order of the decision ratios, and obtaining the data with the largest decision ratio among the real-time collected data, whose decision ratio is E_max;

[0117] S704, calculate the similarity between the real-time collected data and the k kinds of instructions generated by the historical intelligent device, and the calculation formula is:

[0118]

[0119] In the formula, ρ is the similarity between the real-time collected data and each instruction generated by the historical intelligent device, n is the number of data corresponding to each instruction generated by the intelligent device in the history, and E h is the data decision ratio corresponding to each instruction generated by the intelligent device in history, h ranges from 1 to n, f is the number of data with the same data type compared with the data corresponding to the instructions generated by the historical intelligent device in the real-time collected data, and w is the number of data with the same data value compared with the data corresponding to the instructions generated by the historical intelligent device in the real-time collected data; E_s_b z Characterize each instruction generated by the intelligent device in history;

[0120] S705: Calculate the k instructions generated by the smart device in history in sequence according to the above formula to obtain k similarities {ρ 1 , 2 , 3 ... k} 1 , 2 , 3 ... k is the calculated similarity of the 1st, 2nd, 3rd, ...kth data, where k is a positive integer;

[0121] S706. After calculating the similarity set, the similarities in the set are compared, and the instruction corresponding to the maximum similarity is selected as the instruction Q generated after the delayed data exists. The intelligent device outputs the instruction Q and issues an early warning for the delayed data.

[0122] If data anomalies or delays occur in smart devices, the smart devices may not be able to generate instructions normally, or may generate erroneous instructions. Therefore, the above method first determines whether the delayed data changes the data, and then generates instructions through similarity. This allows the smart device to continue to generate instructions and work after an anomaly occurs. If the data is less affected, it only needs to issue an early warning for the delayed data, which will not affect the work of the smart device. When it is determined that the delayed data will change the instructions generated by the smart device, the smart device will stop working immediately and issue an early warning to avoid greater losses caused by erroneous instructions generated by the smart device.

[0123] An equipment control and management system based on AI artificial intelligence, the equipment control and management system includes a data collection module, a threshold calculation module, a feature extraction module, a judgment and early warning module, a similarity calculation module and an instruction output module;

[0124] The data collection module is used to collect historical instructions and corresponding data generated by smart devices, and to collect data detected by sensors in smart devices in real time;

[0125] The threshold calculation module is used to calculate the decision ratio threshold and the standard time threshold according to the instructions and data generated by the intelligent device in history;

[0126] The feature extraction module is used to calculate the data decision ratio corresponding to the instructions generated by the intelligent device in the collected history, and obtain the features of different instructions;

[0127] The judgment and early warning module is used to judge the real-time collected data according to the calculated decision ratio threshold and standard time threshold;

[0128] The similarity calculation module is used to calculate the similarity between the real-time collected data and each instruction after determining the delay data, and generate the corresponding instruction according to the similarity;

[0129] The instruction output module is used to generate instructions through calculation, and then have the intelligent device output the generated instructions and issue an early warning for delayed data.

[0130] The threshold calculation module includes a standard time threshold calculation unit and a decision ratio threshold calculation unit;

[0131] The standard time threshold calculation unit is used to collect the time required for the smart device to generate instructions k times in history, calculate the average time value of the collected k times required for the smart device to generate instructions, and set the average time value as the standard time threshold for the smart device to generate instructions;

[0132] The decision ratio threshold calculation unit is used to collect k different instructions generated by the intelligent device in history, compare the data in each instruction, and when the instruction changes, extract the changed data set in the changed instruction, analyze the data set, calculate the total decision ratio of the data in the k instructions, and then calculate the average value, and set the average value as the data decision ratio threshold when the instruction changes;

[0133] The judgment and early warning module includes a delay data judgment unit and a decision ratio judgment unit;

[0134] The delayed data judgment unit is used to collect the transmission time of different data collected by different sensors in real time, and compare the transmission time of real-time data with the standard time threshold to determine whether there is delayed data;

[0135] The decision ratio judgment unit is used to monitor the instructions generated by the smart device in real time, and after determining that there is delay data, extract the decision ratio of the delay data, compare the decision ratio of the delay data with the decision ratio threshold, and determine whether the delay data will change the instructions generated by the smart device. When it is determined that the delay data will change the instructions generated by the smart device, the instruction feature similarity is calculated based on each data collected in real time, and the corresponding instructions are generated; when it is determined that the delay data will not change the instructions generated by the smart device, the smart device stops generating instructions and issues an early warning, prompting the staff of the smart device abnormality.

[0136] Example 1

[0137] When collecting data from a smart fire alarm device at a certain location, after collecting and analyzing the data generated by the historical instructions of the smart fire alarm device, it is found that the smart device has three types of data, namely temperature, smoke, and humidity, and two instructions can be generated based on the three data, namely, water spraying fire extinguishing {temperature, smoke}, and alarm prompt {smoke, humidity}. After calculation, the decision ratio of humidity is 40%, the decision ratio of temperature is 50%, and the decision ratio of smoke is 20%; and the decision ratio threshold of the smart device is calculated to be 30%, and the standard time threshold is 1.5 seconds; the characteristics of water spraying fire extinguishing and alarm prompt are 40% and 50% respectively;

[0138] The smart device is now monitored in real time, and the transmission time of the temperature, humidity, and smoke data is 0.5s, 1s, and 2s respectively. The transmission time of the three data is compared with the standard time threshold, and it is determined that the transmission time of the smoke data is greater than the standard transmission time, and the smoke data is delayed data;

[0139] It is determined that the smoke detection sensor in the intelligent fire extinguishing alarm device at the location is abnormal, resulting in a slow transmission time of the smoke data, and then it is determined whether the smoke data will affect the generation of the alarm device command; specifically:

[0140] The smoke data is judged to determine whether the smoke data will affect the generation of the smart device instructions. If 20% < 30%, it is determined that the smoke data will not affect the generation of the smart device instructions.

[0141] Exclude smoke data from real-time data, and calculate the similarity between the excluded data obtained from real-time monitoring and the sprinkler fire extinguishing and alarm prompts. First, after comparison, the number of data types that are the same as the real-time data and the sprinkler fire extinguishing is 1, the number of data values ​​that are the same is 1, the number of data types that are the same as the real-time data and the alarm prompt is 1, and the number of data values ​​that are the same is 0. The formula is:

[0142]

[0143] The calculated similarity between real-time data and water sprinkler fire extinguishing is 2.91, and the similarity between real-time data and alarm prompt is 2.31. Comparing the calculated similarities, it is found that water sprinkler fire extinguishing > alarm prompt. Therefore, it is finally calculated that the instruction generated by the smart device after this data delay should be water sprinkler fire extinguishing.

[0144] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0145] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A device control management method based on AI artificial intelligence, characterized by: The method comprises the following steps: S100, collecting different data collected by different sensors when different instructions are generated in the history of the smart device, analyzing the different data of each instruction collected from the different instructions, and obtaining a decision ratio of each data when the smart device generates different instructions; The specific steps to obtain the decision-making ratio of different data to generate instructions for smart devices are: S101, marking the data corresponding to different instructions generated in the history of the smart device, and generating a data set S as {s1, s2, s3...s n }, s1, s2, s3...s n are the 1st, 2nd, 3rd...nth types of data in different instructions, and n is a positive integer; suppose the total number of different instructions collected is k, and the data in the k instructions are collected and marked, and finally k different data sets are generated, and each data in each data set is different; S102, after collecting the data corresponding to different instructions in the smart device, compare all the data sets corresponding to the different instructions, calculate the number of times p of each data change when the instructions are different, sort each data from small to large according to the calculated number of changes, and obtain the number of changes of each data after sorting as {p1, p2, p3...p n }, p1, p2, p3...p n is the number of changes of the data in the 1st, 2nd, 3rd...nth order after sorting, where n is a positive integer; S103, calculating the decision ratio of each data according to the number of changes of each data in different instructions, the formula is: E in the formula i is the decision ratio of the data in the ith order, p i is the number of changes of the i-th data in all different instructions, and k is the number of all different instructions generated historically; S200, comparing and analyzing different instructions generated in the history of the smart device, calculating the total decision ratio of all data that have changed when the instruction changes according to the decision ratios of different data, and finally obtaining the data decision ratio threshold when the instruction changes; S300, after calculating the data decision ratio threshold when the instruction changes, calculate the standard time threshold when the smart device generates the instruction according to the time records of different data transmissions when the smart device generates the instruction in history; S400, analyzing all instructions generated in the history of the smart device to obtain different features of different instructions, and generating an instruction feature set including all instruction features; S500, there is a central processor in the smart device. When the smart device is working, the central processor generates different instructions after receiving different data, monitors the transmission time of data collected by different sensors to the central processor in real time, determines whether the data is delayed according to the standard time threshold, and obtains delayed data; S600, when it is determined that the data is delayed, determining whether the delayed data causes the instruction generated by the smart device to change according to the decision ratio threshold; S700. When it is determined that the delayed data will not change the instructions generated by the smart device, the similarity between the real-time collected data and each different instruction in the history is calculated based on the instruction feature set, and the instruction with the greatest similarity is selected as the output instruction through comparison, and an early warning is issued for the delayed data.

2. According to claim 1, a device control management method based on AI artificial intelligence is characterized by: The specific steps of finally obtaining the data decision ratio threshold when the instruction changes in S200 are as follows: S201, collect k different instructions generated by the smart device in history, and obtain different instruction sets {x1, x2, x3...x k }, x1, x2, x3...x k The instructions generated for the 1st, 2nd, 3rd, ... kth smart devices collected, where k is a positive integer; S202, compare the data in each instruction, when the instruction changes, extract the changed data set in the changed instruction as {s_b1, s_b2, s_b3...s_b z }, s_b1, s_b2, s_b3...s_b z is the 1st, 2nd, 3rd...zth data that changes when the instruction changes, and z is a positive integer; S203, obtain the changed data set {s_b1, s_b2, s_b3...s_b z } and the decision ratio of each data to calculate the total decision ratio of all changed data, the formula is: In the formula, E_a is the total decision ratio of all data that change when the instruction changes, and E r is the decision ratio of each type of data that changes, r takes values ​​from 1 to z; S204. Calculate the total data decision ratio in k instructions, then calculate the average value, and set the average value as the data decision ratio threshold E_y when the instruction changes.

3. The device control management method based on AI artificial intelligence according to claim 1 is characterized by: The specific steps of calculating the standard time threshold for data transmission required for the smart device to generate an instruction in S300 are: S301, collecting the time required for the smart device to generate instructions k times in history, specifically {t1, t2, t3...t k }, t1, t2, t3...t k The time required to generate the instruction for the 1st, 2nd, 3rd, ...kth time, where k is a positive integer; S302 , calculating the average time value of the time required for the smart device to generate instructions for the k times collected, and setting the average time value as the standard time threshold t_y for the smart device to generate instructions.

4. The device control management method based on AI artificial intelligence according to claim 1 is characterized by: The specific steps of obtaining different features of different instructions in S400 are: S401, collecting all instructions generated by the smart device in history and classifying the instructions, classifying the same instructions into the same category, and obtaining the different instruction sets in S201; S402, using the change data set {s_b1, s_b2, s_b3...s_b z }, sort the change data from small to large according to the decision ratio of each data in the change data set, and suppose the decision ratio of the change data in each instruction change is sorted as {E_s_b1, E_s_b2, E_s_b3...E_s_b z }, E_s_b1, E_s_b2, E_s_b3...E_s_b z is the decision ratio of the changed data in the 1st, 2nd, 3rd...zth order in each instruction change, where z is a positive integer; the data type with the largest decision ratio among the changed data sets in the changed instruction is selected as the feature E_s_b of the corresponding instruction z ; S403, calculate the features of k different instructions, and generate an instruction feature set {E_s_b z1 、E_s_b z2 、E_s_b z3 ...E_s_b zk }, E_s_b z1 、E_s_b z2 、E_s_b z3 ...E_s_b zk are the characteristics of the 1st, 2nd, 3rd...kth instructions, where k is a positive integer.

5. The device control management method based on AI artificial intelligence according to claim 1 is characterized by: The specific steps of obtaining the delay data in S500 are: S501, real-time acquisition of the transmission time of different data collected by different sensors, assuming that the collected data transmission time is {t_s1, t_s2, t_s3...t_s n }, t_s1, t_s2, t_s3...t_s n is the transmission time of the 1st, 2nd, 3rd...nth data collected in real time, where n is a positive integer; S502, determine when t_s i >t_y, there is a delay in data transmission time, let the delayed data s_d be the data s corresponding to the transmission time i , the value of i is from 1 to n.

6. The device control management method based on AI artificial intelligence according to claim 1 is characterized by: The specific steps of determining whether the delay data causes the instruction generated by the smart device to change in S600 are as follows: S601, when real-time monitoring is performed on instructions generated by the smart device, it is determined that there is delayed data s_d; S602. The decision ratio of the extracted delay data is E_d. When it is judged that E_d≤E_y, the delay data will not change the instructions generated by the smart device; when it is judged that E_d>E_y, the delay data will change the instructions generated by the smart device, and then the smart device stops generating instructions and issues an early warning to prompt the staff that the smart device is abnormal.

7. The device control management method based on AI artificial intelligence according to claim 1 is characterized by: The specific steps of calculating the similarity between the real-time collected data and each different instruction in the history in S700 are: S701, after determining that the delay data will not change the instruction generated by the smart device, sort the real-time collected data, exclude the calculated delay data, and use the remaining data to calculate the similarity between the collected data and each instruction. Specifically, the remaining data is {s_u1, s_u2, s_u3...s_u j }, s_u1, s_u2, s_u3...s_u j The first, second, third, ...jth data collected in real time after excluding delayed data, where j is a positive integer; S702, compare the real-time collected data set with the data corresponding to the instructions generated by the smart device in the history, calculate the number of data with the same data type as f, then compare the value of each data, and set the number of data with the same data value as w; S703, extracting the decision ratios corresponding to the real-time collected data, sorting the real-time collected data in ascending order of the decision ratios, and obtaining the data with the largest decision ratio among the real-time collected data, whose decision ratio is E_max; S704, calculate the similarity between the real-time collected data and the k kinds of instructions generated by the historical intelligent device, and the calculation formula is: In the formula, ρ is the similarity between the real-time collected data and each instruction generated by the historical intelligent device, n is the number of data corresponding to each instruction generated by the intelligent device in the history, and E h is the data decision ratio corresponding to each instruction generated by the intelligent device in history, h ranges from 1 to n, f is the number of data with the same data type compared with the data corresponding to the instructions generated by the historical intelligent device in the real-time collected data, and w is the number of data with the same data value compared with the data corresponding to the instructions generated by the historical intelligent device in the real-time collected data; E_s_b z Characterize each instruction generated by the intelligent device in history; S705: Calculate the k instructions generated by the smart devices in history in sequence according to the above formula to obtain k similarities {ρ1, ρ2, ρ3...ρ k }, ρ1, ρ2, ρ3...ρ k is the calculated similarity of the 1st, 2nd, 3rd, ...kth data, where k is a positive integer; S706. After calculating the similarity set, the similarities in the set are compared, and the instruction corresponding to the maximum similarity is selected as the instruction Q generated after the delayed data exists. The intelligent device outputs the instruction Q and issues an early warning for the delayed data.

8. An AI-based equipment control management system using an AI-based equipment control management method according to any one of claims 1 to 7, characterized in that: The equipment control and management system includes a data collection module, a threshold calculation module, a feature extraction module, a judgment and early warning module, a similarity calculation module and an instruction output module; The data collection module is used to collect historical instructions and corresponding data generated by smart devices, and to collect data detected by sensors in smart devices in real time; The threshold calculation module is used to calculate the decision ratio threshold and the standard time threshold according to the instructions and data generated by the intelligent device in history; The feature extraction module is used to calculate the data decision ratio corresponding to the instructions generated by the intelligent device in the collected history, and obtain the features of different instructions; The judgment and early warning module is used to judge the real-time collected data according to the calculated decision ratio threshold and standard time threshold; The similarity calculation module is used to calculate the similarity between the real-time collected data and each instruction after determining the delay data, and generate the corresponding instruction according to the similarity; The instruction output module is used to generate instructions through calculation, and then have the intelligent device output the generated instructions and issue an early warning for delayed data.

9. The AI-based equipment control and management system according to claim 8, characterized in that: The threshold calculation module includes a standard time threshold calculation unit and a decision ratio threshold calculation unit; The standard time threshold calculation unit is used to collect the time required for the smart device to generate instructions k times in history, calculate the average time value of the collected k times required for the smart device to generate instructions, and set the average time value as the standard time threshold for the smart device to generate instructions; The decision ratio threshold calculation unit is used to collect k different instructions generated by the intelligent device in history, compare the data in each instruction, and when the instruction changes, extract the changed data set in the changed instruction, analyze the data set, calculate the total decision ratio of the data in the k instructions, and then calculate the average value, and set the average value as the data decision ratio threshold when the instruction changes; The judgment and early warning module includes a delay data judgment unit and a decision ratio judgment unit; The delayed data judgment unit is used to collect the transmission time of different data collected by different sensors in real time, and compare the transmission time of real-time data with the standard time threshold to determine whether there is delayed data; The decision ratio judgment unit is used to monitor the instructions generated by the smart device in real time, and after judging the existence of delay data, extract the decision ratio of the delay data, compare the decision ratio of the delay data with the decision ratio threshold, and judge whether the delay data will change the instructions generated by the smart device. When it is judged that the delay data will change the instructions generated by the smart device, the instruction feature similarity is calculated according to each data collected in real time, and the corresponding instruction is generated; When it is determined that the delayed data will not change the instructions generated by the smart device, the smart device stops generating instructions and issues an early warning, prompting the staff member that the smart device is abnormal.

Images