阅读说明：本技术 一种智能药盒系统及工作方法 (Intelligent medicine box system and working method ) 是由 王琰 于 2021-05-27 设计创作，主要内容包括：本发明提出了一种智能药盒,具体是一种智能药盒系统及工作方法,属于智能药盒领域；智能药盒系统,包括：主控模块、电源模块、天线模块、开关模块、存储模块,GPRS模块、语音模块和震动模块；本发明最主要的功能就是提醒用户按时服药；该药盒小巧便携,并且开发了语音提醒等智能化服务,每次的服药情况包括正常服药、少服、超时服药等都会被记录并上传,以便下次做到更为人性化的服务；同时本发明在进行云计算任务调度,采用粒子群算法,同时加入惯性权重防止调度时出现的局部收敛,更好的减少云计算调度的服务时间,解决了寻找最优解的困难；在改变惯性权重的基础上加入随机粒子个体,降低粒子群全部出现在最优粒子周围的情况。(The invention provides an intelligent medicine box, in particular to an intelligent medicine box system and a working method, belonging to the field of intelligent medicine boxes; an intelligent cartridge system, comprising: the system comprises a main control module, a power supply module, an antenna module, a switch module, a storage module, a GPRS module, a voice module and a vibration module; the main function of the invention is to remind the user to take the medicine on time; the medicine box is small and portable, intelligent services such as voice reminding and the like are developed, and the medicine taking condition of each time including normal medicine taking, less medicine taking, overtime medicine taking and the like can be recorded and uploaded, so that more humanized services can be realized next time; meanwhile, the cloud computing task scheduling is carried out, the particle swarm algorithm is adopted, and meanwhile, the inertial weight is added to prevent local convergence during scheduling, so that the service time of the cloud computing scheduling is better reduced, and the difficulty in finding the optimal solution is solved; random particle individuals are added on the basis of changing the inertia weight, and the situation that the particle swarm is totally around the optimal particles is reduced.)

1. An intelligent cartridge system, comprising:

the main control module is used for broadcasting and processing the information in real time;

the power supply module is used for providing power support for the intelligent medicine box;

the antenna module is used for establishing communication connection with other equipment;

the switch module is used for switching on and off the intelligent medicine box;

the storage module is used for storing the working data of the intelligent medicine box;

the GPRS module is used for carrying out communication between the intelligent medicine box and the server;

the voice module is used for broadcasting and reminding the user to take the medicine on time

And the vibration module is used for asking for help when the user encounters danger.

2. The intelligent medicine box system of claim 1,

the vibration module comprises a gravity sensor, the SDA and the SCL of the gravity sensor are respectively a data line and a clock line, and the bidirectional two-wire system bus is in serial communication with the Bluetooth module; when the medicine box receives external force, the gravity sensor sends a signal to the Bluetooth module, so that the medicine box is recovered to a working state from a sleeping state, and communication is established in the mobile phone; meanwhile, the gravity sensor is provided with a threshold value on the coordinate axis, after a user carrying the medicine box falls down accidentally, the value of the sensor in the coordinate axis direction changes and exceeds the set threshold value, the sensor outputs a signal to the Bluetooth module, and after receiving the signal, the Bluetooth module uploads the situation by utilizing the GPRS, so that the rescue can be timely obtained.

3. The intelligent medicine box system of claim 2,

the switch module comprises a Hall switch, the level state output by the Hall switch is related to the opening and closing state of the medicine box, and when the switch module is closed, the Hall switch outputs a low level because the Hall switch and the magnet are in an attraction state; on the contrary, when the medicine box is opened, a high level is output, and the level signal can enable the main control chip to time; if the timing is more than 10s, opening the box to take the medicine; if the time is less than 10s, the operation is error and no record is made.

4. The intelligent medicine box system of claim 1,

the power supply module further comprises

A charging circuit: the device is used for automatically stopping running when the voltage of the battery is the maximum value, and finishing charging when the voltage of the battery is the preset voltage value during charging so as to finish charging;

the reverse current protection circuit is used for preventing misoperation such as reverse connection of a power supply from damaging hardware equipment, the voltage stabilizer is used for protecting reverse current, and when the power supply is connected reversely, the charging module does not work, so that charging can be carried out only when the power supply is correctly connected.

5. An operating method of an intelligent medicine box is characterized by comprising the following steps:

collecting data;

the cloud server processes the data;

convenient intelligent processing.

6. The method of claim 5, wherein the intelligent medicine box is provided with a plurality of medicine boxes,

firstly, data acquisition: the intelligent medicine box can record the medicine taking condition of the user and upload information to the cloud server, and the information data can be stored, analyzed and the like to judge the medicine taking state of the user and the condition at the moment.

7. The working method of the intelligent medicine box according to the claim 5,

the cloud server performs data processing: after the server receives the uploaded data, the data can be stored, analyzed and the like, the state of the user can be reasonably judged at the moment, and if an accident situation occurs, the server can timely send information to a mobile phone of a designated contact person for online reminding.

8. The method of claim 5, wherein the intelligent medicine box is provided with a plurality of medicine boxes,

convenient and intelligent; the shared reliable network service is combined with the monitoring system of the Internet of things, the condition of the device can be inquired anytime and anywhere based on an application program operated by a webpage, and in addition, data received by the sensor is imaged and sent to a user, and the body condition of the user can also be observed; in addition, the application program can be developed in a computer and can also be developed in a mobile phone, so that convenience is brought to users.

9. The method of claim 5, wherein the intelligent medicine box is provided with a plurality of medicine boxes,

convenient intelligent processing includes: a cloud computing task scheduling method;

the cloud computing task scheduling specifically comprises the following steps:

submitting a task;

decomposing task data;

after the task scheduler receives the subtasks, the task scheduler automatically checks the number of the virtual resources, and then combines the virtual resources and the subtasks for processing.

10. The method of claim 9, wherein the intelligent medicine box is provided with a plurality of medicine boxes,

after the task scheduler receives the subtasks, the random particle individuals are added according to the inertia weight added in the particle algorithm, the situation that all particle swarms appear around the optimal particles is reduced, the random particle individuals are added, and the situation that all particle swarms appear around the optimal particles is reduced.

Technical Field

The invention provides an intelligent medicine box, in particular relates to an intelligent medicine box system and a working method, and belongs to the field of intelligent medicine boxes.

Background

In recent years, the aging process of China is gradually accelerated, a series of social problems of the old people need to be concerned, and the health guarantee of the old people is more and more emphasized. Elderly people cannot well cope with daily normal life due to memory deterioration and physical function reduction, and have more and more chronic diseases. Many elderly people need to take medicines frequently to relieve the disease, so auxiliary medical tools are also on the market. In the report issued by the national health administration, more and more people with chronic diseases such as diabetes, hypertension, hyperlipidemia, heart thrombus and the like can be seen, the disease proportion is synchronously increased along with the increase of the age, one of two people has chronic diseases on average, and the old needs to take corresponding medicines for a long time to treat the chronic diseases. The old aged forgets to take the medicine due to the problems of memory decline and the like, so that the curing effect is poor, and the physical health condition is influenced to cause the occurrence of complications. Therefore, various auxiliary intelligent medicine boxes are gradually invented to solve the problem. However, the existing intelligent medicine box still has a plurality of problems, such as too single type of stored medicine, too simple function of medicine taking reminding, and inconvenient use of a part of old people. In the product background, it is indispensable to design a more intelligent and more efficient medicine box system. Due to the high morbidity of chronic diseases, the intelligent medicine box has the function of timely reminding, and the medicine taking habits of the old are recorded and feedback suggestions are recommended by utilizing the Internet of things, so that the experience optimization is achieved. The medical effect is better improved while the information of each old person is pertinently recorded, and the pressure of tracking treatment by medical staff is reduced.

Meanwhile, in the prior art, the cloud computing is adopted for task scheduling during task scheduling, the cloud computing service provided by a cloud computing provider can make the cloud computing provider obtain benefits, the cost performance of the product is guaranteed, and when the cloud computing provider is used by a user, positive feedback is brought to a merchant by the high cost performance, so that the cloud computing is promoted to be continuously developed. In order to provide high-quality cloud computing service, a cloud computing provider uses Qos to evaluate the quality of a virtual network service provided by cloud computing, so that the use satisfaction degree of a user is considered in a cloud computing research scope. Only if the system has the characteristics of excellent service quality and short-time task completion, resources can be utilized to a greater extent, and users can be better served. Moreover, load balance of cloud computing needs to be realized, with the great increase of resource usage, cloud computing load is easy to generate an unbalance phenomenon, once the load is unbalanced, the function of the whole system is necessarily influenced, and therefore the load balance of cloud computing is also necessary to be researched.

Disclosure of Invention

The purpose of the invention is as follows: an intelligent medicine box system and a working method are provided to solve the problems in the prior art.

The technical scheme is as follows: an intelligent medicine box system and a working method thereof comprise:

the main control module is used for broadcasting and processing the information in real time;

the power supply module is used for providing power support for the intelligent medicine box;

the antenna module is used for establishing communication connection with other equipment;

the switch module is used for switching on and off the intelligent medicine box;

the storage module is used for storing the working data of the intelligent medicine box;

the GPRS module is used for carrying out communication between the intelligent medicine box and the server;

the voice module is used for broadcasting and reminding the user to take the medicine on time

And the vibration module is used for asking for help when the user encounters danger.

Preferably, the vibration module comprises a gravity sensor, the SDA and the SCL of the gravity sensor are respectively a data line and a clock line, and the bidirectional two-wire bus is in serial communication with the Bluetooth module; when the medicine box receives external force, the gravity sensor sends a signal to the Bluetooth module, so that the medicine box is recovered to a working state from a sleeping state, and communication is established in the mobile phone; meanwhile, the gravity sensor is provided with a threshold value on the coordinate axis, after a user carrying the medicine box falls down accidentally, the value of the sensor in the coordinate axis direction changes and exceeds the set threshold value, the sensor outputs a signal to the Bluetooth module, and after receiving the signal, the Bluetooth module uploads the situation by utilizing the GPRS, so that the rescue can be timely obtained.

Preferably, the switch module comprises a hall switch, the level state output by the hall switch is related to the opening and closing state of the medicine box, and when the switch module is closed, the hall switch outputs a low level because the hall switch and the magnet are in an attraction state; on the contrary, when the medicine box is opened, a high level is output, and the level signal can enable the main control chip to time; if the timing is more than 10s, opening the box to take the medicine; if the time is less than 10s, the operation is error and no record is made.

Preferably, the power supply module further comprises

A charging circuit: the device is used for automatically stopping running when the voltage of the battery is the maximum value, and finishing charging when the voltage of the battery is the preset voltage value during charging so as to finish charging;

the reverse current protection circuit is used for preventing misoperation such as reverse connection of a power supply from damaging hardware equipment, the voltage stabilizer is used for protecting reverse current, and when the power supply is connected reversely, the charging module does not work, so that charging can be carried out only when the power supply is correctly connected.

An operating method of an intelligent medicine box comprises the following steps:

collecting data;

the cloud server processes the data;

convenient intelligent processing.

Preferably, the data acquisition is performed firstly: the intelligent medicine box can record the medicine taking condition of the user and upload information to the cloud server, and the information data can be stored, analyzed and the like to judge the medicine taking state of the user and the condition at the moment.

Preferably, the cloud server performs data processing: after the server receives the uploaded data, the data can be stored, analyzed and the like, the state of the user can be reasonably judged at the moment, and if an accident situation occurs, the server can timely send information to a mobile phone of a designated contact person for online reminding.

Preferably, the method is convenient and intelligent; the shared reliable network service is combined with the monitoring system of the Internet of things, the condition of the device can be inquired anytime and anywhere based on an application program operated by a webpage, and in addition, data received by the sensor is imaged and sent to a user, and the body condition of the user can also be observed; in addition, the application program can be developed in a computer and can also be developed in a mobile phone, so that convenience is brought to users.

Preferably, the convenient intelligent processing comprises: a cloud computing task scheduling method;

the cloud computing task scheduling specifically comprises the following steps:

submitting a task;

decomposing task data;

after the task scheduler receives the subtasks, the task scheduler automatically checks the number of the virtual resources, and then combines the virtual resources and the subtasks for processing.

Preferably, after receiving the subtask, the task scheduler adds the random particle individuals according to the inertia weight added in the particle algorithm, so as to reduce the situation that the particle swarm all appears around the optimal particle and reduce the situation that the particle swarm all appears around the optimal particle by adding the random particle individuals. .

Has the advantages that: the intelligent medicine box is suitable for the forgetful old people and the busy people, and has the main function of reminding the user to take medicine on time. The medicine box is small and portable, intelligent services such as voice reminding and the like are developed, and the medicine taking condition of each time including normal medicine taking, less medicine taking, overtime medicine taking and the like can be recorded and uploaded, so that more humanized services can be realized next time; meanwhile, the cloud computing task scheduling is carried out, the particle swarm algorithm is adopted, and meanwhile, the inertial weight is added to prevent local convergence during scheduling, so that the service time of the cloud computing scheduling is better reduced, and the difficulty in finding the optimal solution is solved; random particle individuals are added on the basis of changing the inertia weight, and the situation that the particle swarm is totally around the optimal particles is reduced.

Drawings

Fig. 1 is a schematic diagram of an intelligent medicine box of the present invention.

Fig. 2 is a circuit diagram of a main control module of the intelligent medicine box system of the present invention.

Fig. 3 is a schematic diagram of an intelligent medicine box system charging circuit and a reverse current protection circuit according to the present invention.

Fig. 4 is a circuit diagram of a GPRS module of the intelligent medicine box system of the present invention.

Fig. 5 is a circuit diagram of a vibration module of the intelligent medicine box system of the present invention.

Fig. 6 is an audio decoding circuit diagram of the intelligent medicine box system of the present invention.

Fig. 7 is a circuit diagram of a warning light of the intelligent medicine box system of the invention.

Fig. 8 is a circuit diagram of the switch module of the intelligent medicine box system of the present invention.

Fig. 9 is a schematic diagram of the operation of the intelligent medicine box system of the present invention.

Fig. 10 is a schematic diagram of the intelligent medicine box system according to the present invention.

FIG. 11 is a cloud computing task framework diagram of the present invention.

FIG. 12 is a cloud computing system architecture diagram of the present invention.

Fig. 13 is a flowchart illustrating a task scheduling process of cloud computing according to the present invention.

FIG. 14 is a graph showing the relationship between inertial weight and population evolution algebra under the LDW strategy of the present invention.

FIG. 15 is a graph of the relationship between the inertial weight and population evolution algebra under the LDW strategy for increasing constant disturbance of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

In one embodiment, as shown in FIG. 1, we propose an intelligent medication box comprising;

the main control module is used for broadcasting and processing the information in real time;

the power supply module is used for providing power support for the intelligent medicine box;

the antenna module is used for establishing communication connection with other equipment;

the switch module is used for switching on and off the intelligent medicine box;

the storage module is used for storing the working data of the intelligent medicine box;

the GPRS module is used for carrying out communication between the intelligent medicine box and the server;

the voice module is used for broadcasting and reminding the user to take the medicine on time

And the vibration module is used for reminding a user.

In one embodiment, the intelligent medicine box is powered by a battery, the power supply voltage is 4.2V, the battery can be charged repeatedly, and the antenna used by the intelligent medicine box is 2.4GHz, so that the Bluetooth can be ensured to communicate in channels with the frequency of 2402 plus 2480MHz, and the normal operation of 39 channels can be maintained; meanwhile, the intelligent medicine box is also provided with a pi-type matching circuit around the Bluetooth module;

specifically, the circuit diagram is shown in fig. 2; the main control module has the advantages of low power consumption and small size, the mobility of the equipment can be greatly increased, the service life of the product is prolonged, and in addition, the real-time performance of the equipment is improved due to strong computing capability. As can be seen, the chip pins 22 are connected to the ceramic antenna, which enables the signal strength to be increased while extending the communication distance. And a pi-type circuit of a capacitor and an inductor is added at the same time, so that filtering can be performed.

In one embodiment, the power module of the intelligent medicine box can repeatedly charge and supply power to the battery, as shown in fig. 3. When the voltage of the battery is 3.2V, the equipment can automatically stop running, the charging voltage during charging is 5V, and when the voltage of the battery is 4.2V, the charging is finished, and the charging is finished at the moment. In order to avoid misoperation such as reverse connection of a power supply and the like to damage hardware equipment, the circuit integrates the voltage stabilizer, so that the reverse current protection effect can be realized, when the power supply is connected reversely, the charging module does not work, and the charging can be ensured to be carried out only when the power supply is correctly connected.

In one embodiment, the GPRS module shown in fig. 4 is used to facilitate the communication between the intelligent medicine box and the server, including the transmission and reception of data and operation signals; in addition, the data such as the medicine taking times, the interval time and the like of a user when using the medicine box can be stored and uploaded to the server, then the medicine taking times, the interval time and the like can be directly set on the server, meanwhile, the modified new data is sent to the medicine box on line, and the medicine box can work according to the new setting; the communication protocol here uses the URAT protocol.

In one embodiment, the vibration module adopts a gravity sensor, the interface communication protocol adopts an I2C protocol, the protocol is composed of two lines, as shown in fig. 5, the SDA and SCL in the figure are a data line and a clock line, respectively, and the bidirectional two-wire bus performs serial communication with the bluetooth module; when the medicine box receives external force, the gravity sensor can send a signal to the Bluetooth module, so that the medicine box recovers to the working state from the sleep state, and communication is established in the mobile phone. In addition, gravity sensor has set up the threshold value on the coordinate axis, and after the user that carries the medicine box had happened the accident and fell down, the value of sensor in the coordinate axis direction can change and surpass the threshold value that sets up, and the sensor can export a signal to bluetooth module this moment, receives this signal after, bluetooth module can utilize GPRS with the condition to upload, just so can in time obtain the succour.

In one embodiment, in addition, the voice module is provided with an audio decoding circuit, the intelligent medicine box can be remotely operated through a mobile phone, after the medicine box is communicated with the mobile phone, a user can send the times of taking medicine, the interval time and the like to the medicine box, in addition, a recorded voice memo can be sent to the medicine box for storage, and when the medicine needs to be taken, the medicine box can release the stored voice to remind the user. The voice broadcast system needs an audio decoder, the schematic diagram is shown in fig. 6, and the module and the bluetooth module adopt an SPI synchronous communication bus, so that the communication is simple.

In one embodiment, the medicine box is further provided with a prompting lamp, fig. 7 is a schematic diagram of three-color LED lamps, wherein the three-color LED lamps are a blue lamp, a green lamp and a red lamp respectively, and are connected in a common anode mode. The three colored lights represent three different situations in which the kit is located. In the process of Bluetooth matching with the mobile phone, the blue lamp is on, and is automatically turned off after matching is successful for 5 s; the red light is on to indicate that the electric quantity is insufficient and the charging is required in time; when the medicine box is charged, when the red light is turned off and the green light is turned on, the charging is finished. When the user waits for the preset medicine taking time, the red LED lamp flickers, if the user takes medicine after the preset time, the red LED lamp is recorded as wrong medicine taking, and the lamp light does not flick at the moment. In addition, if the medicine box is opened in an abnormal state and the opening time exceeds 10min, the LED lamp emits light to remind the user of the abnormal state until the medicine box is closed.

In one embodiment, the switch module includes a hall switch, as shown in fig. 8. The switch is used for distinguishing whether the medicine is normally taken or not by opening the box, and the principle is as follows: the output level state of the Hall switch in the box is related to the opening and closing state of the medicine box, and when the Hall switch is closed, the Hall switch can output low level because the Hall switch and the magnet are in an attraction state; conversely, when the medicine box is opened, a high level is output, and the level signal enables the main control chip to time. If the timing is more than 10s, opening the box to take the medicine; if the time is less than 10s, the operation is error and no record is made.

As shown in fig. 9, the main function of the intelligent medicine box is to remind the user to take medicine on time. The medicine box is small and portable, intelligent services such as voice reminding and the like are developed, and the medicine taking condition of each time including normal medicine taking, less medicine taking, overtime medicine taking and the like can be recorded and uploaded, so that more humanized services can be realized next time; the medicine box realizes normal communication with the mobile phone based on the Bluetooth chip, and data can be transmitted by using the chip within 100m, so that the effective distance is longer and more convenient. When the preset time is reached, the medicine box can play the voice prompt 10 minutes in advance, and then the medicine box can start timing until forty minutes passes. Within the forty minutes, the medicine taking reminding can be broadcasted every 5 minutes, and if the medicine is taken before the timing is finished, the medicine is taken on time; if the medicine box is opened after the timing is finished, the medicine box is recorded as the wrong medicine taking; if the medicine is not taken all the time, the medicine is regarded as forgotten to be taken.

As shown in fig. 10, the intelligent medicine box is a high-tech product integrating various module functions, and covers knowledge of various subjects such as power electronics, automatic control principle, embedded system, sensor control, and the like; the medicine taking condition of the user every time can be uploaded to the cloud end by the medicine box for storage, and then is sent to the mobile phone of each appointed person, such as a parent, a private doctor and the like. The work flow of the medicine box is as follows: when the battery begins to supply power, the bluetooth chip resumes work, and before reaching the time of taking medicine of predetermineeing, can be to signals such as LED lamp, audio player, the medicine box can begin to remind the user to take medicine this moment to upload and save the condition of taking medicine. The medicine box utilizes the bluetooth chip to realize the communication with the cell-phone, sends the user who stores and takes medicine the condition, receives the new order that the user set up.

The invention explains the defects of the current medical industry and points out the advancing direction for further promoting the modernization and intelligent development of the medical industry, and the integration of the Internet of things and the cloud computing technology into the medical industry is an important breakthrough point, which can help to establish a more intelligent and higher-safety medical monitoring system, and the two high-tech technologies can provide more high-speed and convenient data processing services.

An operating method of an intelligent medicine box comprises the following steps:

collecting data;

the cloud server processes the data;

convenient intelligent processing.

Further, data acquisition is performed firstly: the intelligent medicine box can record the medicine taking condition of the user and upload information to the cloud server, and the information data can be stored, analyzed and the like to judge the medicine taking state of the user and the condition at the moment;

secondly, the cloud server processes data: after the server receives the uploaded data, the data can be stored, analyzed and the like, the state of the user is reasonably judged at the moment, and if an accident situation occurs, the server can timely send information to a mobile phone of a designated contact person for online reminding;

finally, convenient and intelligent operation is realized; the shared reliable network service is combined with the monitoring system of the internet of things, the condition of the device can be inquired anytime and anywhere based on the application program operated by the webpage, and in addition, the data received by the sensor is imaged and sent to the user, and the body condition of the user can also be observed. In addition, the application program can be developed in a computer and can also be developed in a mobile phone, so that convenience is brought to users.

As shown in fig. 11, the sensing layer is the layer closest to the user in the internet of things system, the intelligent medicine box is the sensing layer in the system, and the sensing layer lays a foundation for the operation of the application layer by collecting the treatment record and daily behavior of the user; the information collected by the sensing layer is communicated through a network layer, the network layer is mainly responsible for data transmission, and the communication protocol adopts http and mqtt protocols of TCMP; the support layer is mainly responsible for the construction of the running environment and mainly is the design of software and hardware of the system; the application layer is a resource interface in the Internet of things system, and human-computer interaction can be realized by compiling a programming program; the sensing layer intelligent gateway is vital, not only is connected with a sensor network and a computer network, but also can regulate and control devices in the system, and the regulation and control of the devices can be started only after the system sends a remote instruction or a local program and the instruction or the program passes verification;

the invention also includes a database; the database is a database without architecture design, and data stored in the database is stored in a K-V attribute mode; because a plurality of sensors are needed under the platform of the internet of things, the attribute of the data model is not fixed, in order to solve the problems, a Dynamo DB database is adopted to store the data, the storage efficiency of the data is high, and the platform of the internet of things calls the data simply and quickly; the database can input, inquire and share data structures of stored contents in various forms, so that the problem caused by repeated data input is effectively solved, meanwhile, the database has sharing performance, each user in the platform can have rights to access the database, and the user has rights to store, share and access other users; the database has a simple structure and is convenient to use. Because the database has sharing property, repeated data storage can be effectively avoided, and the elaboration of a large amount of complex databases is avoided; in addition, the data content is safe and reliable. The database requires that the data with the same content stored in the database need to be consistent, and self-maintenance can be started to protect the data when the database runs in a background.

A cloud computing task scheduling algorithm, comprising:

submitting a task;

decomposing task data;

after the task scheduler receives the subtasks, the task scheduler automatically checks the number of the virtual resources, and then combines the virtual resources and the subtasks for processing.

In one embodiment, as shown in fig. 12, the specific steps of the cloud computing task scheduling process are as follows:

firstly, a user submits a task, the cloud user purchases a server from a service provider providing the cloud server, the servers are virtually stored in a network, a common user can use resources in the server without mastering the working principle of the server, the use threshold of the server is greatly reduced, and the service provider provides the resources for the purchasing user to complete the task;

secondly, the user needs to decompose the task after submitting the task, which is to adopt a distributed method to decompose the task, as can be seen from fig. 10, the user does not need to have specific cognition on the processing after submitting the task, because of the distributed principle, the system decomposes the task submitted by the user, and after obtaining a plurality of small tasks, the small tasks are received by a task scheduler and then the next work is carried out;

and finally, after the task scheduler receives the subtasks, the task scheduler can automatically check the number of the virtual resources and then combine the virtual resources and the subtasks for processing.

In the above, the cloud computing service provided by the cloud computing provider ensures the cost performance of the product while enabling the user to gain income, and when the user uses the cloud computing provider, the cost performance brings forward feedback to the merchant, so that the cloud computing is promoted to develop forward and continuously. In order to provide high-quality cloud computing service, a cloud computing provider uses Qos to evaluate the quality of a virtual network service provided by cloud computing, so that the use satisfaction degree of a user is considered in a cloud computing research scope. Only if the system has the characteristics of excellent service quality and short-time task completion, resources can be utilized to a greater extent, and users can be better served. Moreover, load balance of cloud computing needs to be realized, with the great increase of resource usage, cloud computing load is easy to generate an unbalance phenomenon, once the load is unbalanced, the function of the whole system is necessarily influenced, and therefore the load balance of cloud computing is also necessary to be researched.

Cloud computing itself is based on virtualization, which does not need to provide real resources to users, but rather connects actual devices to a virtual network through task scheduling. Compared with distributed computing, cloud computing has the capability of managing resources in a distributed mode, and has the characteristics of virtualized resources, multi-thread programming and the like. With the wide application of cloud computing to various fields, the precision and efficiency of task scheduling in cloud computing must be ensured. Aiming at the task scheduling problem, the task scheduling algorithm for the PSO serves as a core of task scheduling;

the task scheduling method of the peer-to-peer network is another method in distributed computing, a new task scheduling mode can be obtained by combining P2P with Grid computing, and a P2P-Grid model can realize the connection among multiple devices while combining tasks and resources; the P2P task scheduling mode places different devices in equal positions, each device can initiate task request and respond to other tasks, and the P2P-Grid task scheduling mode has the capability of peer-to-peer response of the devices. P2P and grid computing can be combined together also because they have many similarities, namely distributed computing and the same dynamic features; the larger scale Grid system is decomposed into many smaller Grid systems by the P2P-Grid model, which are the same level and all of which are one of the most important cores. And the sub-Grid system can request tasks and respond to tasks, that is, Grid-Peer of the sub-system can perform distributed work with other sub-systems. In P2P-Grid, after the user finishes the task submission, the task manager and the resource manager receive the submitted task and send the submitted task into the network resource group, the resource monitor and the resource scheduler continuously and circularly call the resource in the network resource group for processing, and finally the processed result is returned to the client to finish the task scheduling.

Furthermore, distributed computing can subdivide a large and complex problem into a plurality of small tasks and then process the small tasks on different resources, and the resource is operated and scheduled through a resource manager, so that the problem of low efficiency in the task scheduling process can be easily caused by the constantly changing resource tasks. Compared with distributed computing, in the cloud computing, a huge cloud resource pool is created by a system, centralized execution of resources at a task center is realized, measures of distributed management greatly improve the execution efficiency of an Internet of things platform, the scheduling process is different from the task scheduling of the distributed computing by using a virtualization technology, and a task scheduling method of the cloud computing is researched; therefore, the characteristics of the cloud computing task scheduling are known.

As shown in fig. 10, which is a cloud computing task scheduling framework diagram, a resource scheduler may receive a task uploaded by a cloud user; and the resource scheduler analyzes the scale of the uploading task to judge the number of the required physical machines. The tasks uploaded by the users can be stored in the cloud Portal, and the resource manager calls the tasks from the tasks. After the physical resources are determined, the VMMonitor synthesizes the physical machines into a virtual machine to complete task scheduling, so that mapping between physical hardware and virtual resources is formed, a user can obtain resources in a resource pool as required, and each user has the same resources in a right distribution pool.

From the above description, we can find that the task scheduler is always dynamically changed, and the addition of a new virtual machine and the exit of an error-reported virtual machine are simultaneous; meanwhile, the number of users in cloud computing is extremely large, so that the tasks of the cloud server are also extremely large, the hardware resources of the cloud server are challenged, and the real-time performance of the task processing can be ensured only by ensuring more hardware devices so as to cope with the continuous increase of the tasks; task scheduling in cloud computing is heterogeneous, and the task scheduling is applied to a heterogeneous platform; the service quality of cloud computing and the time consumed for completing tasks are the characteristics of achieving economy if high-efficiency and high-quality cloud computing is achieved at the same time.

Different computing nodes exist in the cloud computing process, and the computing performances corresponding to the different computing nodes are different; the same task can be operated on only one node; one node cannot simultaneously perform two or more tasks of calculation; in practice, cloud computing resource scheduling can be regarded as a process for solving an integer programming problem; the time required by the node to complete the task can be pre-judged by evaluating the capacity of the node for executing the million instructions within 1s, so that the time required by the node to complete the task is pre-judged after the node receives the task, and the scheduling problem is optimized; optimizing to reduce the time for the nodes to complete tasks, and establishing a mathematical model of a cloud computing resource scheduling problem:

in the formula (I), the compound is shown in the specification,a set of tasks is represented that is,is the number of tasks;a set of tasks is represented that is,the total number of the nodes is the total number of the nodes,in order to execute the execution of the time,comparing the time required for each task to complete among all nodes in order to execute the maximum execution time, wherein the maximum task time is set toWhen is coming into contact withWhen the ith task operates at the jth node, parameters are calculated(ii) a If not operating, is；

Specific maximum task execution timeIs defined as follows:

to reduce the time for the nodes to complete the task, the optimization objective function is converted into:

in the formula (I), the compound is shown in the specification,the minimum time required to complete a task;

cloud resource scheduling involves integer programming, for which case, as defined aboveThe cloud resource scheduling method can constitute the cloud resource scheduling vector into a vector, and the vector plays a role in decision in cloud resource scheduling and is also called as a decision variable; if the cloud resource data needing to be scheduled is too large, the calculation difficulty is increased, and the whole solving process of the cloud resource scheduling problem can be regarded as a process for solving a complicated NP problem. For such problems, an ordinary algorithm such as a newton method cannot be used for calculation, and a more advanced intelligent algorithm is used for solving the problems. Nowadays, the problem of optimized resource scheduling is solved by applying an intelligent algorithm in more applications, and the method is better at presentThe algorithm includes tabu search algorithm, ant colony algorithm, etc. However, the above intelligent algorithm is easy to fall into local optimization in the iterative operation process, and this disadvantage affects the solution of parameters in the whole problem re-optimization process. The particle swarm algorithm is also an intelligent algorithm applied to solving the optimization problem, can effectively solve the defects, is more convenient and faster than the former algorithm, has strong parameter optimization capability, is applied to a complex nondeterministic polynomial problem, and can well improve the interference of local optimization;

assuming that the number of particle groups is N, these particles exist in a space of D dimension, and the positions of the particles are vectorized in this space, which can be described asSimultaneously vectorizing the spatial movement velocity of the particles, and recording asThe individual extremum is recorded asThe extreme value of all the particles in the best state is recordedThat is, the algorithm finds the optimal value, and in the basic algorithm, the calculation formula required by the continuous update iteration is as follows:

in the formula (I), the compound is shown in the specification,n is the serial number of the particle,is a particle ofThe ratio of vitamin to vitamin is,for the number of iterations of the algorithm,andexpressed as a constant, the value of which generally ranges from；Is a real number randomly existing on (0, 1);

adding inertial weight coefficients to the above formulaTherefore, the optimization performance of the algorithm is effectively improved, and the problem that the algorithm falls into local optimization is avoided; thus, an updated iterative relationship is obtained:

in the formula, the convergence performance depends on the inertia factorThe size of (2) can not only enable the particles to continuously maintain the original motion habit, but also enlarge the search space of the algorithm.The larger the global search is, the shorter the time required for the global search is, and the time taken for the optimization becomes short, but it is difficult to find a correct solution;the smaller the size, the more beneficial it is to find and calculate the correct solution locally, but the time spent in the optimization becomes longer, and even more so, the optimization process encounters local optimality. Therefore, before the particle swarm algorithm is used, it is necessary to select an appropriate one so that the search time and accuracy of the algorithm are appropriate.

Specifically, the relevant parameters in the algorithm have great influence on the convergence of the algorithm, so that the particle swarm optimization mainly improves the approaching speed at the beginning of the population development according to the characteristic of the particle swarm optimization relevant parameters; in the later algorithm, the high-precision solution is obtained, so that the convergence performance of the algorithm can be improved by setting the inertia weight to be a dynamically changed characteristic; because the space structure where the particle swarm is located is complicated and nonlinear, the improvement degree of the LDW strategy on the particle swarm algorithm in the prior art is limited; so if some perturbations are added in the optimization process, whose values are fixed constants, these perturbations are encountered during the implementation of the LDW strategy,the method can be increased, and when the algorithm cannot find a better current optimal solution, the measures can effectively avoid the face algorithm from falling into local optimization, and continue to enter all the spaces for searching to find global optimization; therefore, the updating iteration relation based on the algorithm for increasing the constant disturbance parameter is as follows:

in the formula (I), the compound is shown in the specification,the inertia weight is a fixed value under the condition of disturbance, and under different disturbance conditions,and in the rest cases,(ii) a The disturbance probability value is set to 0.1 after comprehensive consideration;

further, the algorithm improves the parameters when in operation, so that the parameters reach an optimized state, and if one of all the particles can reach the optimal state and reaches the best position close to the optimal solution, other particles also approach the position under the optimal solution; if the position is locally best, then the algorithm cannot solve for the extreme value in the current position, at which point the state of premature convergence occurs. Therefore, the greatest disadvantage is that in the algorithm optimization process, the particles occasionally find the local optimal solution, although the method provided in the previous subsection can effectively improve the disadvantage, the iterative mode of the algorithm is not fundamentally changed, and the particle swarm algorithm is still easy to gather around the optimal position, so that the particle swarm algorithm lacks diversity;

thereby utilizing the above formulaThe value of (c) is adjusted as the environment changes, even if it is dynamically adaptive; based on the LDW particle swarm optimization algorithm based on the addition of constant disturbance parameters,make itWith a varying coefficientMultiplying the obtained product by the coefficient within 0.9-1.1 to obtain a product for replacing the original inertia weight coefficient, i.e.(ii) a Therefore, the algorithm updating iteration relation after the dynamic self-adaptive parameters are introduced is as follows:

therefore, the invention provides the inaccuracy of algorithm optimization caused by the lack of diversity in the original algorithm, and the inaccuracy is solved by increasing the randomness of individual particles. According to the update method of the particle swarm algorithm, as the number of iterations increases, the optimal value appears, so the particles can be gathered at the position, and the single result lacks diversity. The introduced random particles have 30 percent of probability to be exchanged with particle objects in the original population, so that the diversity is compensated by the measures, and the low-probability replacement mode cannot change the algorithm optimizing direction.

Based on the analysis, the invention introduces dynamic adaptive parameters and random individual particles, improves the LDW strategy based on constant disturbance parameters, and provides a novel improved particle swarm algorithm to solve the problem of cloud resource scheduling. In order to verify the effectiveness and the correctness of the method, the improved particle swarm algorithm is compared with a particle swarm algorithm and a common particle swarm algorithm based on the addition of constant disturbance parameters; the following were used:

the optimal solution is 0.0 by applying the De jong function. The specific De jong function is formulated as follows:

as can be seen from fig. 13, the algorithm proposed by the present invention requires a small number of iterations to find an optimal solution, and the optimization effect is the best among the three methods, and it can be seen that the approximately optimal solution is:(ii) a Function value of optimal solution。

The Schaffer function, the optimal solution 0.0, is as follows:

fig. 14 shows that the method proposed by the present invention has the best test result, and the obtained approximately optimal solution is:value of the optimal solution function0。

Therefore, the particle swarm algorithm is introduced for solving; because the basic particle swarm algorithm cannot well search the optimal solution for the large task scheduling, lacks robustness to diversity and is easy to fall into the local optimal condition, the improved particle swarm algorithm is provided on the basis; constant disturbance and random individual particles are added, so that local optimization can be avoided, and the problems of particle aggregation and deficient algorithm diversity can be effectively solved. Finally, the feasibility and the effectiveness of the method are verified by using a test function and a cloud resource scheduling example, the method has a better optimization effect from the test result, and a searched high-precision optimization result is very suitable for solving the cloud resource scheduling problem and can be widely applied to solving the problem.

As noted above, while the present invention has been shown and described with reference to certain preferred embodiments, it is not to be construed as limited thereto. Various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.