Property management system of main control chip

Abstract

The invention discloses a property management system of a main control chip, which comprises the main control chip, a data acquisition module, a data storage module, a data analysis module, a data processing module, a dynamic time sequence priority scheduling module and a data statistics archiving module, according to the deep learning event hidden danger model, identifying hidden danger events in a community; the data processing module forms a solution of event processing through algorithm processing of hidden events; the dynamic time sequence priority scheduling module is used for generating a dynamic time sequence priority scheduling model; according to the invention, for the problem of low efficiency of property management and security event processing, a hidden event is rapidly identified through a deep learning event hidden event model, and a hidden event processing scheme is rapidly found through a cosine similarity algorithm and a whale algorithm; and main control chip resource scheduling required by each happening hidden danger event is solved through a dynamic time sequence priority scheduling model, so that the hidden danger event can be quickly solved.

Description

Technical Field

[0001] The present invention belongs to the field of chips for property management, and specifically refers to a property management system with a main control chip. Background Art

[0002] With the acceleration of urbanization, in order to improve the efficiency and service quality of property management, more and more property management companies have begun to adopt intelligent property management systems. Intelligent property management systems include sensors, control equipment, data processing centers and other components installed in various areas of the property.

[0003] Chinese Patent Publication No.: CN117196541A discloses a property management monitoring and management system and method, including an abnormal data collection module, an abnormal data processing module, a task management module and a maintenance management module. The monitoring and management method of the present invention collects abnormal data of several monitoring devices in the data monitoring subsystem of the property management center, and analyzes the historical data of the monitoring points corresponding to the abnormal data, so as to determine the processing method of the abnormal data of the monitoring points corresponding to the abnormal data according to whether the abnormal rate of the corresponding monitoring points exceeds the standard, and generates corresponding maintenance tasks according to the abnormal data under the corresponding processing method through the task allocation unit, and determines the skill proficiency and number of maintenance personnel of the maintenance tasks, so as to achieve precise control of the property management process;

[0004] Chinese Patent Publication No.: CN114663257A discloses a property management monitoring and management system, which relates to the field of monitoring and management technology, including a call system, a monitoring system, an event response system, an emergency event self-inspection system, an emergency plan handling opinion information storage library, an event emergency management distribution system and a response speed tracking system; the system of the present invention can achieve efficient and rapid response to emergency events occurring in the property management area of ​​a community, solves the problems of slow response speed and unreasonable personnel allocation in the traditional event response mechanism, can effectively improve the efficiency and quality of property monitoring management, has obvious benefits in improving property service capabilities, and can effectively reduce the harmful effects of emergency events, thereby ensuring the safety of life and property of community residents.

[0005] However, current products mainly focus on the precise control and convenience of the property management process. The property management system in the existing technology is still relatively slow in identifying community events and is unable to quickly propose solutions to potential danger events. The system's efficiency in handling potential danger events is still relatively low, and the main control chip cannot handle multiple potential danger events efficiently and quickly at the same time. Summary of the invention

[0006] In order to solve the problems in the above-mentioned prior art that the property management system is still slow in identifying community events, cannot quickly propose solutions to potential danger events, the system is still low in efficiency in handling potential danger events, and the main control chip is unable to handle multiple potential danger events efficiently and quickly, the present invention proposes a property management system with a main control chip to improve the above-mentioned problems.

[0007] The specific application is as follows:

[0008] A property management system of a main control chip, comprising a main control chip, a data acquisition module, a data storage module, a data analysis module, a data processing module, a dynamic time sequence priority scheduling module and a data statistics archiving module;

[0009] The main control chip is equipped with a data acquisition module, a data storage module, a data analysis module, a data processing module, a dynamic timing priority scheduling module and a data statistics archiving module. The data acquisition module is connected to the data storage module, the data storage module is connected to the data analysis module, the data analysis module is connected to the data processing module, and the data processing module is connected to the data statistics archiving module;

[0010] The data acquisition module is used to collect historical and real-time related data in the property management system, collect image data through the camera, and collect voice data through the voice recognizer. The data is stored in the data storage module;

[0011] The data storage module is used to store the data collected by the data collection module;

[0012] The data analysis module is used to segment different objects in the data acquisition module, segment each object into a first object and a first additional object, input feature information of the first object and the first additional object into a deep learning event hidden danger model, calculate event hidden danger coefficients for the first object and the first additional object according to the deep learning event hidden danger model, obtain a first event hidden danger coefficient, activate a first trajectory instruction if the first event hidden danger coefficient is greater than a first preset event hidden danger coefficient, track the first object and the first additional object according to the first trajectory instruction, generate a second event hidden danger coefficient, and generate corresponding target event warning information according to the second event hidden danger coefficient;

[0013] The data processing module is used to process the target event warning information into a characteristic vector to generate a characteristic vector, wherein the target event warning information is the event warning information corresponding to the security event that is occurring, and the cosine similarity algorithm and the whale algorithm are used to match the security event that is occurring to search for the security event that has occurred in the data storage module, extract features from the security event that has occurred, and perform reference analysis based on the extracted features to obtain a solution to the security event that is occurring;

[0014] The dynamic timing priority scheduling module generates a dynamic timing priority scheduling model based on the data acquisition module, the data storage module, the data analysis module, the data processing module and the data statistics archiving module. The dynamic timing priority scheduling model is used to solve the resource scheduling of the main control chip required for each security event that is occurring;

[0015] The data statistics archiving module is used to record the process of the security incident, including recording using voice, text and images, and saving the recorded results in the data storage module.

[0016] Further, the data storage module includes a data backup unit and a data recovery unit;

[0017] The data backup unit is used to regularly back up data to prevent data loss;

[0018] The data recovery unit is used to quickly recover data when the data is lost or damaged.

[0019] Furthermore, it is characterized in that the data analysis module includes an object segmentation submodule, a coefficient calculation submodule, a trajectory submodule, a tracking submodule and a safety identification submodule;

[0020] The segmentation object submodule is used to segment the image collected by the data acquisition module according to the target segmentation object model embedded in the main control chip, and determine the first object and the first additional object, wherein the first additional object is the user object of the first object, and the first object and the first additional object include adult-electric vehicle, adult-pet, child-pet, child-child riding vehicle;

[0021] The coefficient calculation submodule is used to input the feature information of the first object and the feature information of the first additional object into the deep learning event hidden danger model, and calculate the event hidden danger coefficients of the first object and the first additional object according to the deep learning event hidden danger model to obtain the first event hidden danger coefficient;

[0022] The trajectory submodule is used to activate the first trajectory instruction for tracking when the first event hidden danger coefficient is greater than the first preset event hidden danger coefficient;

[0023] The tracking submodule is used to track the first object and the first additional object according to the first trajectory instruction to obtain a plurality of tracking frames;

[0024] The safety identification submodule is used to perform feature identification based on the tracking multi-frame images, output a first tracking hidden danger coefficient, and generate event warning information when the first tracking hidden danger coefficient is greater than a second preset event hidden danger coefficient.

[0025] Furthermore, it is characterized in that the coefficient calculation submodule also includes:

[0026] Acquiring characteristic information of the first object, including body shape characteristics, gender characteristics, and age characteristics;

[0027] Acquire feature information of the first additional object, including additional volume features, static and dynamic features, and constraint features;

[0028] The feature information of the first object and the feature information of the first additional object are used as training data and input into the pre-trained deep learning event potential risk model for calculation to obtain a first event potential risk coefficient that identifies the degree of event potential risk of the first object and the first additional object.

[0029] Furthermore, it is characterized in that the method for performing feature recognition according to the tracking multiple frames of images and outputting a first tracking hidden danger coefficient specifically comprises:

[0030] Tracking the behavior trajectories of the first object and the first additional object respectively to obtain a first tracking trajectory and a second tracking trajectory;

[0031] Performing fitness identification according to the first tracking trajectory and the second tracking trajectory to obtain a trajectory deviation fitness and a trajectory time fitness;

[0032] Obtaining a trajectory comprehensive fitness degree according to the trajectory deviation fitness degree and the trajectory time fitness degree;

[0033] If the comprehensive fitness of the trajectory does not meet the preset fitness threshold, a trajectory event hidden danger coefficient is generated;

[0034] According to the tracking multiple frames of images, the dynamic relationship between the behaviors of the first object and the first additional object is associated and identified to obtain a dynamic event hidden danger coefficient;

[0035] According to the trajectory event hidden danger coefficient and the dynamic event hidden danger coefficient, a first tracking hidden danger coefficient is output.

[0036] Furthermore, it is characterized in that the data processing module includes a feature vector submodule, a cosine algorithm submodule, a whale algorithm submodule and a fusion processing submodule;

[0037] The feature vector submodule is used to perform vector processing on the event warning information through vector engineering;

[0038] The cosine algorithm submodule searches all historical data stored in the data storage module through a cosine similarity algorithm to find the event warning information that has occurred and has the highest similarity to the event warning information, and extracts the security event feature vector corresponding to the event warning information that has occurred;

[0039] Cosine similarity algorithm:

[0040]

[0041] The cosine similarity algorithm can be used to calculate the similarity of two security events. The larger the cosθ value, the more similar the two security events are, and vice versa. i Represents the set of feature vectors corresponding to the security events that are occurring, y i A set of feature vectors corresponding to security events that have occurred;

[0042] The cosine similarity algorithm processing process:

[0043] M1: Feature vector construction, using feature engineering to extract feature vectors from ongoing security events and all security events that have occurred and stored in the data storage module;

[0044] M2: Cosine similarity calculation, using the cosine similarity algorithm to calculate the security event that is currently occurring and all the security events that have occurred to be searched, and setting the calculation result to Q;

[0045] M3: Similarity evaluation. The larger the value of cosθ is, the more similar it is. A threshold is set to D. It is determined whether the Q value is greater than the threshold D. If it is greater than the threshold D, the security event class that has occurred corresponding to the security event that is occurring is found. The security event class that has occurred is a security event of type I, and the number is at least greater than 1.

[0046] The whale algorithm submodule is used to search for solutions to ongoing security incidents, and the solutions include the manpower required to solve the security incidents, the material resources required to solve the security incidents, and the optimal solution process for solving the security incidents, wherein the manpower includes the most suitable personnel for solving the security incidents, and the material resources include the most suitable material resources for solving the security incidents;

[0047] Whale algorithm implementation process:

[0048] D=|CX*(t)-X(t)|,

[0049] X(t+1)=X*(t)-AD,

[0050] t is the current iteration number, A and C are the coefficients, X*(t) represents the best position so far, that is, the optimal solution, and X(t) represents the current search position;

[0051] A=2ar 1 -a,

[0052] C=2r 2 ,

[0053] a=2-2t / Tmax,

[0054] Among them, r 1 and r 2 is a random number in (0, 1), the value of a decreases linearly from 2 to 0, t represents the current number of iterations, Tmax represents the maximum number of iterations,

[0055] X(t+1)=X * (t)+D * *e bd *cos(2πd),

[0056] Among them, D * =|X * (t)-X(t)|, represents the distance between the location of the current security event to be occurred and the location of the security event that has been searched, X * (t) is the best position vector so far, b is a constant used to define the shape of the spiral, and d is a random number in (-1, 1).

[0057] D=|C*X rand -X(t)|,

[0058] X(t+1)=X rand -A*D,

[0059] Among them, X rand is the randomly selected position vector of the currently occurring security incident. The algorithm sets that when A≥1, the number of security incidents that have been explored is set to N, the boundary of the security incidents that have occurred is set to the N-month security incident records of the data storage module, the number of iterations is set to T, and the optimal feature set is finally searched to obtain the corresponding manpower required to solve the currently occurring security incident, the corresponding material resources required to solve the currently occurring security incident, and the optimal solution process corresponding to the solution of the currently occurring security incident. The search condition is the optimal solution process corresponding to the security incident solution.

[0060] The fusion processing submodule is used to fuse the results of the cosine algorithm submodule and the whale algorithm submodule to generate a solution corresponding to the current security incident. The fused data includes the time point, manpower, material resources and optimal solution process of the security incident. The fused data generates a data flow overview by adopting a graph theory algorithm, which is used for property managers to view the progress of security incident processing in real time.

[0061] Furthermore, the specific implementation process of the dynamic timing priority scheduling module is as follows:

[0062] L1: Based on the timing control function of the main control chip, a heap sorting algorithm is used to construct a binary heap structure to sort the tasks of each module to establish a priority queue, and dynamically adjust the tasks according to their urgency and importance to generate a task priority queue;

[0063] L2: Based on the task priority queue, a linear programming algorithm is used to design and implement a strategy for resource allocation of the main control chip, and the simplex method is used to allocate resources, balance task demand and resource supply, and generate a resource allocation model;

[0064] L3: Based on the resource allocation model, a decision tree algorithm is used to build classification rules and regression models to analyze processing requirements, dynamically adjust task priorities and main control chip resource allocation, and generate an adjusted task scheduling strategy;

[0065] L4: Based on the adjusted task scheduling strategy, a reinforcement learning algorithm is used to continuously adapt to changes in processing requirements by implementing a reward feedback mechanism and state space analysis, optimize task priorities and main control chip resource configuration, and generate a dynamic timing priority scheduling model. The main control chip resources include the number of chip cores, the number of chip threads, the chip cache, and the chip power consumption.

[0066] Further, it is characterized in that the data statistics archiving module includes an event personnel voice information entry submodule and an event process information entry submodule;

[0067] The event personnel voice information recording submodule is used to translate and recognize the human voice, obtain the semantics of the translated human voice, and convert the semantics into text records;

[0068] The event process information entry submodule is used to record in detail the entire process of the security incident that occurred, including the cause of the security incident, the security incident resolution process, the security incident processing results, and the manpower and material resources required in the security incident resolution process and the detailed processing flow.

[0069] The beneficial effects of the property management system of a main control chip of the present invention are as follows:

[0070] The present invention collects historical and real-time related data in the property management system through a data acquisition module, collects image data through a camera, and collects voice data through a voice recognizer; the data storage module is used to store the data collected by the data acquisition module; the data analysis module performs segmentation processing on different objects in the data acquisition module, divides each object into a first object and a first additional object, inputs the feature information of the first object and the first additional object into a deep learning event hidden danger model, calculates the event hidden danger coefficient of the first object and the first additional object according to the deep learning event hidden danger model, obtains the first event hidden danger coefficient, if the first event hidden danger coefficient is greater than the first preset event hidden danger coefficient, activates the first trajectory instruction, tracks the first object and the first additional object according to the first trajectory instruction, generates the second event hidden danger coefficient, and generates corresponding target event warning information according to the second event hidden danger coefficient; the data processing module performs characterization vector processing on the target event warning information to generate a feature vector, and the target event warning information is the target event hidden danger coefficient that is being generated. The event warning information corresponding to the security event that has occurred is matched with the cosine similarity algorithm and the whale algorithm to search for the security event that has occurred in the data storage module, and the feature extraction is performed through the security event that has occurred. The reference analysis is performed through the extracted features to obtain the solution to the security event that is occurring; the dynamic timing priority scheduling module generates a dynamic timing priority scheduling model, and the dynamic timing priority scheduling model is used to solve the resource scheduling of the main control chip required for each security event that is occurring; the data statistics archiving module records the process of the security event that has occurred, including recording using voice, text and images, and saving the recorded results in the data storage module; the present invention aims at the problem of inefficient property management in handling security events, quickly identifies potential danger events through a deep learning event potential danger model, quickly finds potential danger event processing solutions through a cosine similarity algorithm and a whale algorithm, and solves the resource scheduling of the main control chip required for each potential danger event that is occurring through a dynamic timing priority scheduling model, so as to achieve rapid resolution of potential danger events. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 This is a system architecture diagram of a property management system of a main control chip of the present invention;

[0072] Figure 2 A data storage module of a property management system of a main control chip of the present invention;

[0073] Figure 3 A data analysis module diagram of a property management system of a main control chip of the present invention;

[0074] Figure 4 This is a data processing module diagram of a property management system of a main control chip of the present invention;

[0075] Figure 5 This is a data statistics and archiving module diagram of a property management system of a main control chip of the present invention.

[0076] Among them, 1-data acquisition module; 2-data storage module; 21-data backup unit; 22-data recovery unit; 3-data analysis module; 31-segmentation object submodule; 32-coefficient calculation submodule; 33-trajectory submodule; 34-tracking submodule; 35-safety identification submodule; 4-data processing module; 41-eigenvector submodule; 42-cosine algorithm submodule; 43-whale algorithm submodule; 44-fusion processing submodule; 5-dynamic timing priority scheduling module; 6-data statistics archiving module; 61-event personnel voice information entry submodule; 62-event process information entry submodule. DETAILED DESCRIPTION

[0077] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. 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.

[0078] Example 1

[0079] Figure 1 This is a system architecture diagram of a property management system with a main control chip according to the present invention.

[0080] The system includes a main control chip, a data acquisition module 1, a data storage module 2, a data analysis module 3, a data processing module 4, a dynamic time sequence priority scheduling module 5 and a data statistics archiving module 6;

[0081] The main control chip is equipped with a data acquisition module 1, a data storage module 2, a data analysis module 3, a data processing module 4, a dynamic timing priority scheduling module 5 and a data statistics archiving module 6. The data acquisition module 1 is connected to the data storage module 2, the data storage module 2 is connected to the data analysis module 3, the data analysis module 3 is connected to the data processing module 4, and the data processing module 4 is connected to the data statistics archiving module 6;

[0082] The data acquisition module 1 is used to collect historical and real-time data in the property management system, collect image data through a camera, and collect voice data through a voice recognizer. The data is stored in the data storage module 2;

[0083] The data storage module 2 is used to store the data collected by the data collection module 1;

[0084] The data analysis module 3 is used to segment different objects in the data acquisition module 1, segment each object into a first object and a first additional object, input feature information of the first object and the first additional object into a deep learning event hidden danger model, calculate event hidden danger coefficients for the first object and the first additional object according to the deep learning event hidden danger model, obtain a first event hidden danger coefficient, activate a first trajectory instruction if the first event hidden danger coefficient is greater than a first preset event hidden danger coefficient, track the first object and the first additional object according to the first trajectory instruction, generate a second event hidden danger coefficient, and generate corresponding target event warning information according to the second event hidden danger coefficient;

[0085] The data processing module 4 is used to process the target event warning information into a characteristic vector to generate a characteristic vector. The target event warning information is the event warning information corresponding to the security event that is occurring. The cosine similarity algorithm and the whale algorithm are used to match the security event that is occurring to search for the security event that has occurred in the data storage module 2, extract features from the security event that has occurred, and perform reference analysis based on the extracted features to obtain a solution to the security event that is occurring.

[0086] The dynamic timing priority scheduling module 5 generates a dynamic timing priority scheduling model based on the data acquisition module 1, the data storage module 2, the data analysis module 3, the data processing module 4 and the data statistics archiving module 6. The dynamic timing priority scheduling model is used to solve the resource scheduling of the main control chip required for each security event that is occurring;

[0087] The data statistics archiving module 6 is used to record the process of the security incident, including recording using voice, text and images, and saving the recorded results in the data storage module 2.

[0088] Further, the data storage module 2 includes a data backup unit 21 and a data recovery unit 22;

[0089] The data backup unit 21 is used to regularly back up data to prevent data loss;

[0090] The data recovery unit 22 is used to quickly recover data when the data is lost or damaged.

[0091] Furthermore, the data analysis module 3 includes an object segmentation submodule 31, a coefficient calculation submodule 32, a trajectory submodule 33, a tracking submodule 34 and a safety identification submodule 35;

[0092] The segmentation object submodule 31 is used to segment the image collected by the data acquisition module according to the target segmentation object model embedded in the main control chip, and determine the first object and the first additional object, wherein the first additional object is the user object of the first object, and the first object and the first additional object include adult-electric vehicle, adult-pet, child-pet, child-child riding vehicle;

[0093] The coefficient calculation submodule 32 is used to input the feature information of the first object and the feature information of the first additional object into the deep learning event hidden danger model, and calculate the event hidden danger coefficients of the first object and the first additional object according to the deep learning event hidden danger model to obtain a first event hidden danger coefficient;

[0094] The trajectory submodule 33 is used to activate the first trajectory instruction for tracking when the first event hidden danger coefficient is greater than the first preset event hidden danger coefficient;

[0095] The tracking submodule 34 is used to track the first object and the first additional object according to the first trajectory instruction to obtain a tracking multi-frame image;

[0096] The safety identification submodule 35 is used to perform feature identification based on the tracking multi-frame images, output a first tracking hidden danger coefficient, and generate event warning information when the first tracking hidden danger coefficient is greater than a second preset event hidden danger coefficient.

[0097] Furthermore, the coefficient calculation submodule 32 also includes:

[0098] Acquiring characteristic information of the first object, including body shape characteristics, gender characteristics, and age characteristics;

[0099] Acquire feature information of the first additional object, including additional volume features, static and dynamic features, and constraint features;

[0100] The feature information of the first object and the feature information of the first additional object are used as training data and input into the pre-trained deep learning event potential risk model for calculation to obtain a first event potential risk coefficient that identifies the degree of event potential risk of the first object and the first additional object.

[0101] Furthermore, the method for performing feature recognition according to the tracking multiple frames of images and outputting a first tracking hidden danger coefficient specifically includes:

[0102] Tracking the behavior trajectories of the first object and the first additional object respectively to obtain a first tracking trajectory and a second tracking trajectory;

[0103] Performing fitness identification according to the first tracking trajectory and the second tracking trajectory to obtain a trajectory deviation fitness and a trajectory time fitness;

[0104] Obtaining a trajectory comprehensive fitness degree according to the trajectory deviation fitness degree and the trajectory time fitness degree;

[0105] If the comprehensive fitness of the trajectory does not meet the preset fitness threshold, a trajectory event hidden danger coefficient is generated;

[0106] According to the tracking multiple frames of images, the dynamic relationship between the behaviors of the first object and the first additional object is associated and identified to obtain a dynamic event hidden danger coefficient;

[0107] According to the trajectory event hidden danger coefficient and the dynamic event hidden danger coefficient, a first tracking hidden danger coefficient is output.

[0108] Furthermore, the data processing module 4 includes a feature vector submodule 41, a cosine algorithm submodule 42, a whale algorithm submodule 43 and a fusion processing submodule 44;

[0109] The feature vector submodule 41 is used to perform vector processing on the event warning information through vector engineering;

[0110] The cosine algorithm submodule 42 searches all historical data stored in the data storage module 2 by matching through the cosine similarity algorithm, finds the event warning information that has occurred with the highest similarity to the event warning information, and extracts the security event feature vector corresponding to the event warning information that has occurred;

[0111] Cosine similarity algorithm:

[0112]

[0113] The cosine similarity algorithm can be used to calculate the similarity of two security events. The larger the cosθ value, the more similar the two security events are, and vice versa. i Represents the set of feature vectors corresponding to the security events that are occurring, y i A set of feature vectors corresponding to security events that have occurred;

[0114] The cosine similarity algorithm processing process:

[0115] M1: Feature vector construction, using feature engineering to extract feature vectors of the ongoing security event and all the security events that have occurred and stored in the data storage module 2;

[0116] M2: Cosine similarity calculation, using the cosine similarity algorithm to calculate the security event that is currently occurring and all the security events that have occurred to be searched, and setting the calculation result to Q;

[0117] M3: Similarity evaluation. The larger the value of cosθ is, the more similar it is. A threshold is set to D. It is determined whether the Q value is greater than the threshold D. If it is greater than the threshold D, the security event class that has occurred corresponding to the security event that is occurring is found. The security event class that has occurred is a security event of type I, and the number is at least greater than 1.

[0118] The whale algorithm submodule 43 is used to search for solutions to ongoing security incidents, wherein the solutions include the manpower required to solve the security incidents, the material resources required to solve the security incidents, and the optimal solution process for solving the security incidents, wherein the manpower includes the most suitable personnel for solving the security incidents, and the material resources include the most suitable material resources for solving the security incidents;

[0119] Whale algorithm implementation process:

[0120] D=|CX*(t)-X(t)|,

[0121] X(t+1)=X*(t)-AD,

[0122] t is the current iteration number, A and C are the coefficients, X*(t) represents the best position so far, that is, the optimal solution, and X(t) represents the current search position;

[0123] A=2ar 1 -a,

[0124] C=2r 2 ,

[0125] a=2-2t / Tmax,

[0126] Among them, r 1 and r 2 is a random number in (0, 1), the value of a decreases linearly from 2 to 0, t represents the current number of iterations, Tmax represents the maximum number of iterations,

[0127] X(t+1)=X * (t)+D * *e bd *cos(2πd),

[0128] Among them, D * =|X * (t)-X(t)|, represents the distance between the location of the current security event to be occurred and the location of the security event that has been searched, X *(t) is the best position vector so far, b is a constant used to define the shape of the spiral, and d is a random number in (-1, 1).

[0129] D=|C*X rand -X(t)|,

[0130] X(t+1)=X rand -A*D,

[0131] Among them, X rand is the randomly selected position vector of the currently occurring security incident. The algorithm sets that when A≥1, the number of security incidents that have been explored is set to N, the boundary of the security incidents that have occurred is set to the N-month security incident records of the data storage module, the number of iterations is set to T, and the optimal feature set is finally searched to obtain the corresponding manpower required to solve the currently occurring security incident, the corresponding material resources required to solve the currently occurring security incident, and the optimal solution process corresponding to the solution of the currently occurring security incident. The search condition is the optimal solution process corresponding to the security incident solution.

[0132] The fusion processing submodule 44 is used to fuse the results of the cosine algorithm submodule and the whale algorithm submodule to generate a solution corresponding to the current security incident. The fused data includes the time point, manpower, material resources and optimal solution process of the security incident. The fused data generates a data flow overview by adopting a graph theory algorithm, which is used for property managers to view the progress of security incident processing in real time.

[0133] Furthermore, the specific implementation process of the dynamic timing priority scheduling module 5 is as follows:

[0134] L1: Based on the timing control function of the main control chip, a heap sorting algorithm is used to construct a binary heap structure to sort the tasks of each module to establish a priority queue, and dynamically adjust the tasks according to their urgency and importance to generate a task priority queue;

[0135] L2: Based on the task priority queue, a linear programming algorithm is used to design and implement a strategy for resource allocation of the main control chip, and the simplex method is used to allocate resources, balance task demand and resource supply, and generate a resource allocation model;

[0136] L3: Based on the resource allocation model, a decision tree algorithm is used to build classification rules and regression models to analyze processing requirements, dynamically adjust task priorities and main control chip resource allocation, and generate an adjusted task scheduling strategy;

[0137] L4: Based on the adjusted task scheduling strategy, a reinforcement learning algorithm is used to continuously adapt to changes in processing requirements by implementing a reward feedback mechanism and state space analysis, optimize task priorities and main control chip resource configuration, and generate a dynamic timing priority scheduling model. The main control chip resources include the number of chip cores, the number of chip threads, the chip cache, and the chip power consumption.

[0138] Furthermore, the data statistics and archiving module 6 includes an event personnel voice information entry submodule 61 and an event process information entry submodule 62;

[0139] The event personnel voice information recording submodule 61 is used to translate and recognize the human voice, obtain the semantics of the translated human voice, and convert the semantics into text records;

[0140] The event process information entry submodule 62 is used to record in detail the entire process of the security event that occurred, including the cause of the security event, the security event resolution process, the security event processing results, and the manpower and material resources required in the security event resolution process and the detailed processing flow.

[0141] Embodiment 2:

[0142] The camera collects image data, and when the collected image data is input into the deep learning event hidden danger model in real time to quickly identify hidden danger events, if a hidden danger event is found, the cosine similarity algorithm and the whale algorithm are used to quickly find the historical record information. The historical record information found is integrated into the human, material resources and optimal solution process required for the current hidden danger event. The dynamic time priority scheduling model is used to solve the main control chip resource scheduling required for each hidden danger event that is occurring, so as to achieve rapid resolution of hidden danger events. After the hidden danger event is processed, the hidden danger event is archived through the data statistics archiving module, including recording using voice, text and images, and the recorded results are saved in the data storage module.

[0143] The beneficial effects of the property management system of a main control chip of the present invention are as follows:

[0144] The present invention collects historical and real-time related data in the property management system through a data acquisition module, collects image data through a camera, and collects voice data through a voice recognizer; the data storage module is used to store the data collected by the data acquisition module; the data analysis module performs segmentation processing on different objects in the data acquisition module, divides each object into a first object and a first additional object, inputs the feature information of the first object and the first additional object into a deep learning event hidden danger model, calculates the event hidden danger coefficient of the first object and the first additional object according to the deep learning event hidden danger model, obtains the first event hidden danger coefficient, if the first event hidden danger coefficient is greater than the first preset event hidden danger coefficient, activates the first trajectory instruction, tracks the first object and the first additional object according to the first trajectory instruction, generates the second event hidden danger coefficient, and generates corresponding target event warning information according to the second event hidden danger coefficient; the data processing module performs characterization vector processing on the target event warning information to generate a feature vector, and the target event warning information is the target event hidden danger coefficient that is being generated. The event warning information corresponding to the security event that has occurred is matched with the cosine similarity algorithm and the whale algorithm to search for the security event that has occurred in the data storage module, and the feature extraction is performed through the security event that has occurred. The reference analysis is performed through the extracted features to obtain the solution to the security event that is occurring; the dynamic timing priority scheduling module generates a dynamic timing priority scheduling model, and the dynamic timing priority scheduling model is used to solve the resource scheduling of the main control chip required for each security event that is occurring; the data statistics archiving module records the process of the security event that has occurred, including recording using voice, text and images, and saving the recorded results in the data storage module; the present invention aims at the problem of inefficient property management in handling security events, quickly identifies potential danger events through a deep learning event potential danger model, quickly finds potential danger event processing solutions through a cosine similarity algorithm and a whale algorithm, and solves the resource scheduling of the main control chip required for each potential danger event that is occurring through a dynamic timing priority scheduling model, so as to achieve rapid resolution of potential danger events.

[0145] The present invention and its embodiments are described above, which is not restrictive. What is shown in the accompanying drawings is only one of the embodiments of the present invention, and the actual content is not limited to this. In short, if ordinary technicians in this field are inspired by it and do not deviate from the purpose of the invention, they can creatively design structural methods and embodiments similar to the technical scheme, which should all fall within the scope of protection of the present invention.

Claims

1. A property management system of a main control chip, characterized in that: It comprises a main control chip, a data acquisition module (1), a data storage module (2), a data analysis module (3), a data processing module (4), a dynamic time sequence priority scheduling module (5) and a data statistics archiving module (6); The main control chip is equipped with a data acquisition module (1), a data storage module (2), a data analysis module (3), a data processing module (4), a dynamic time sequence priority scheduling module (5) and a data statistics archiving module (6); the data acquisition module (1) is connected to the data storage module (2), the data storage module (2) is connected to the data analysis module (3), the data analysis module (3) is connected to the data processing module (4), and the data processing module (4) is connected to the data statistics archiving module (6); The data acquisition module (1) is used to collect historical and real-time related data in the property management system, collect image data through a camera, and collect voice data through a voice recognizer, and the data is stored in the data storage module (2); The data storage module (2) is used to store the data collected by the data collection module (1); The data analysis module (3) is used to segment different objects in the data acquisition module (1), segment each object into a first object and a first additional object, input feature information of the first object and the first additional object into a deep learning event hidden danger model, calculate event hidden danger coefficients for the first object and the first additional object according to the deep learning event hidden danger model, obtain a first event hidden danger coefficient, activate a first trajectory instruction if the first event hidden danger coefficient is greater than a first preset event hidden danger coefficient, track the first object and the first additional object according to the first trajectory instruction, generate a second event hidden danger coefficient, and generate corresponding target event warning information according to the second event hidden danger coefficient; The data processing module (4) is used to process the target event warning information into a characteristic vector to generate a characteristic vector, wherein the target event warning information is the event warning information corresponding to the security event that is currently occurring, and the cosine similarity algorithm and the whale algorithm are used to match the security event that is currently occurring to search for the security event that has already occurred in the data storage module (2), extract features from the security event that has already occurred, and perform reference analysis based on the extracted features to obtain a solution to the security event that is currently occurring; The dynamic time sequence priority scheduling module (5) generates a dynamic time sequence priority scheduling model based on the data acquisition module (1), the data storage module (2), the data analysis module (3), the data processing module (4) and the data statistics archiving module (6), wherein the dynamic time sequence priority scheduling model is used to solve the resource scheduling of the main control chip required for each security event that is occurring; The data statistics archiving module (6) is used to record the process of the security incident, including recording using voice, text and images, and storing the recorded results in the data storage module (2).

2. A property management system of a main control chip according to claim 1, characterized in that: The data storage module (2) comprises a data backup unit (21) and a data recovery unit (22); The data backup unit (21) is used to regularly back up data to prevent data loss; The data recovery unit (22) is used to quickly recover data when the data is lost or damaged.

3. A property management system of a main control chip according to claim 2, characterized in that: The data analysis module (3) comprises an object segmentation submodule (31), a coefficient calculation submodule (32), a trajectory submodule (33), a tracking submodule (34) and a safety identification submodule (35); The object segmentation submodule (31) is used to segment the image collected by the data acquisition module according to the target segmentation object model embedded in the main control chip, and determine the first object and the first additional object, wherein the first additional object is the user object of the first object, and the first object and the first additional object include adult-electric vehicle, adult-pet, child-pet, and child-child riding vehicle; The coefficient calculation submodule (32) is used to input the feature information of the first object and the feature information of the first additional object into the deep learning event hidden danger model, and calculate the event hidden danger coefficients of the first object and the first additional object according to the deep learning event hidden danger model to obtain a first event hidden danger coefficient; The trajectory submodule (33) is used to activate the first trajectory instruction for tracking when the first event hidden danger coefficient is greater than the first preset event hidden danger coefficient; The tracking submodule (34) is used to track the first object and the first additional object according to the first trajectory instruction to obtain multiple tracking image frames; The safety identification submodule (35) is used to perform feature identification based on the tracking multiple frames of images, output a first tracking hidden danger coefficient, and generate event warning information when the first tracking hidden danger coefficient is greater than a second preset event hidden danger coefficient.

4. A property management system of a main control chip according to claim 3, characterized in that: The coefficient calculation submodule (32) further includes: Acquiring characteristic information of the first object, including body shape characteristics, gender characteristics, and age characteristics; Acquire feature information of the first additional object, including additional volume features, static and dynamic features, and constraint features; The feature information of the first object and the feature information of the first additional object are used as training data and input into the pre-trained deep learning event potential risk model for calculation to obtain a first event potential risk coefficient that identifies the degree of event potential risk of the first object and the first additional object.

5. A property management system of a main control chip according to claim 4, characterized in that: The method for performing feature recognition according to the tracking multiple frames of images and outputting a first tracking hidden danger coefficient specifically includes: Tracking the behavior trajectories of the first object and the first additional object respectively to obtain a first tracking trajectory and a second tracking trajectory; Performing fitness identification according to the first tracking trajectory and the second tracking trajectory to obtain a trajectory deviation fitness and a trajectory time fitness; Obtaining a trajectory comprehensive fitness degree according to the trajectory deviation fitness degree and the trajectory time fitness degree; If the comprehensive fitness of the trajectory does not meet the preset fitness threshold, a trajectory event hidden danger coefficient is generated; According to the tracking multiple frames of images, the dynamic relationship between the behaviors of the first object and the first additional object is associated and identified to obtain a dynamic event hidden danger coefficient; According to the trajectory event hidden danger coefficient and the dynamic event hidden danger coefficient, a first tracking hidden danger coefficient is output.

6. A property management system of a main control chip according to claim 5, characterized in that: The data processing module (4) includes a feature vector submodule (41), a cosine algorithm submodule (42), a whale algorithm submodule (43) and a fusion processing submodule (44); The feature vector submodule (41) is used to perform vector processing on the event warning information through vector engineering; The cosine algorithm submodule (42) searches all historical data stored in the data storage module (2) through a cosine similarity algorithm, finds the event warning information that has occurred and has the highest similarity to the event warning information, and extracts the security event feature vector corresponding to the event warning information that has occurred; Cosine similarity algorithm: The cosine similarity algorithm can be used to calculate the similarity of two security events. The larger the cosθ value, the more similar the two security events are, and vice versa. i Represents the set of feature vectors corresponding to the security events that are occurring, y i A set of feature vectors corresponding to security events that have occurred; The cosine similarity algorithm processing process: M1: feature vector construction, using feature engineering to extract feature vectors from the security events that are occurring and all the security events that have occurred and are stored in the data storage module (2); M2: Cosine similarity calculation, using the cosine similarity algorithm to calculate the security event that is currently occurring and all the security events that have occurred to be searched, and setting the calculation result to Q; M3: Similarity evaluation. The larger the value of cosθ is, the more similar it is. A threshold is set to D. It is determined whether the Q value is greater than the threshold D. If it is greater than the threshold D, the security event class that has occurred corresponding to the security event that is occurring is found. The security event class that has occurred is a security event of type I, and the number is at least greater than 1. The whale algorithm submodule (43) is used to search for solutions to ongoing security incidents, wherein the solutions include the manpower required to solve the security incidents, the material resources required to solve the security incidents, and the optimal solution process for solving the security incidents, wherein the manpower includes the most suitable personnel for solving the security incidents, and the material resources include the most suitable material resources for solving the security incidents; Whale algorithm implementation process: D=|CX*(t)-X(t)|, X(t+1)=X*(t)-AD, t is the current iteration number, A and C are the coefficients, X*(t) represents the best position so far, that is, the optimal solution, and X(t) represents the current search position; A=2ar1-a, C＝2r2， a=2-2t / Tmax, Among them, r1 and r2 are random numbers in (0,1), the value of a decreases linearly from 2 to 0, t represents the current number of iterations, and Tmax represents the maximum number of iterations. X(t+1)=X * (t)+D * *e bd *cos(2πd), Among them, D * =|X * (t)-X(t)| represents the distance between the location of the current security event to be occurred and the location of the security event that has been searched for. * (t) is the best position vector so far, b is a constant used to define the shape of the spiral, and d is a random number in (-1,1). D=∣C*X rand -X(t)∣, X(t+1)=X rand -A*D, Among them, X rand is the randomly selected position vector of the currently occurring security incident. The algorithm sets that when A≥1, the number of security incidents that have been explored is set to N, the boundary of the security incidents that have occurred is set to the N-month security incident records of the data storage module, the number of iterations is set to T, and the optimal feature set is finally searched to obtain the corresponding manpower required to solve the currently occurring security incident, the corresponding material resources required to solve the currently occurring security incident, and the optimal solution process corresponding to the solution of the currently occurring security incident. The search condition is the optimal solution process corresponding to the security incident solution. The fusion processing submodule (44) is used to fuse the results of the cosine algorithm submodule and the whale algorithm submodule to generate a solution corresponding to the security incident that is currently occurring. The fused data includes the time point, manpower, material resources and optimal solution process of the security incident. The fused data generates a data flow overview by adopting a graph theory algorithm, which is used for property managers to view the progress of security incident processing in real time.

7. A property management system of a main control chip according to claim 6, characterized in that: The specific implementation process of the dynamic timing priority scheduling module (5) is as follows: L1: Based on the timing control function of the main control chip, a heap sorting algorithm is used to construct a binary heap structure to sort the tasks of each module to establish a priority queue, and dynamically adjust the tasks according to their urgency and importance to generate a task priority queue; L2: Based on the task priority queue, a linear programming algorithm is used to design and implement a strategy for resource allocation of the main control chip, and the simplex method is used to allocate resources, balance task demand and resource supply, and generate a resource allocation model; L3: Based on the resource allocation model, a decision tree algorithm is used to build classification rules and regression models to analyze processing requirements, dynamically adjust task priorities and main control chip resource allocation, and generate an adjusted task scheduling strategy; L4: Based on the adjusted task scheduling strategy, a reinforcement learning algorithm is used to continuously adapt to changes in processing requirements by implementing a reward feedback mechanism and state space analysis, optimize task priorities and main control chip resource configuration, and generate a dynamic timing priority scheduling model. The main control chip resources include the number of chip cores, the number of chip threads, the chip cache, and the chip power consumption.

8. A property management system of a main control chip according to claim 7, characterized in that: The data statistics and archiving module (6) comprises an event personnel voice information input submodule (61) and an event process information input submodule (62); The event personnel voice information recording submodule (61) is used to translate and recognize the human voice, obtain the semantics of the translated human voice, and convert the semantics into text records; The event process information entry submodule (62) is used to record in detail the entire process of the security event that occurred, including the cause of the security event, the security event resolution process, the security event processing results, and the manpower and material resources required in the security event resolution process and the detailed processing flow.

Images