CN107239853A - A kind of wisdom steward system and its method of work based on cloud computing - Google Patents

Abstract

The invention discloses a smart housekeeper system based on cloud computing and its working method. The system includes an enterprise-level client and an ordinary individual client; the enterprise-level client includes a product configuration module, a product registration module, a product Details management module, product association configuration module, portal management module, sales financial management module and courier administrator management module. The common personal client includes a settlement module, a user information module, a housekeeper service module, a data report query module and a data navigation module. The invention adopts the crossover operator of the basic cloud self-adaptive genetic algorithm, and realizes the minimization of the operating cost of the butler management system and the timely and accurate processing of uncertain information. The method based on cloud computing proposed by the present invention improves the speed of optimal solution convergence, increases the search range of the global optimal solution, reduces the running time and cost of the smart housekeeper system, improves the operating efficiency of the system and shortens the response time. time.

Description

A smart housekeeper system based on cloud computing and its working method

technical field

The invention relates to a smart housekeeping system platform and method, in particular to a smart housekeeping system platform and method based on the combination of cloud computing technology and intelligent quantum algorithm, belonging to the technical field of cloud computing technology and Internet application management.

Background technique

With the development of cloud computing and Internet technology, more and more O2O platforms and APP systems based on cloud platforms and Internet have emerged, which bring convenience to people's lives. For example, APP-based commodity procurement management, social security payment management, property fee, water and electricity fee management system, etc. In these application systems, since various types of data are scattered and disorderly, how to effectively collect and reasonably classify user data, and push data information that needs to be managed in a timely manner and track data to users according to the classification results Navigation and updating are the main problems that such systems need to solve.

Most of the housekeeper management system platform designs commonly used in the prior art are designed based on general heuristic algorithms. This type of design method is slow in searching data, the results of cluster analysis are not thorough, and it is difficult to obtain the optimal solution. Moreover, due to the decentralized management of data, it is difficult to Extract regular information from data to effectively predict and provide reference for future behavior. This reduces data management efficiency and weakens the predictive capabilities of data management.

In the application of information management platform based on cloud computing environment, some data is redundant and invalid, and must be processed by appropriate means, such as data cleaning, data denoising, data cluster analysis, etc., according to the type and scale of data , Obtain information such as channels, and carry out retrospective design on data information. Common data processing algorithms in the prior art include: Dijkstra algorithm, genetic algorithm, ant colony algorithm, simulated annealing algorithm and tabu search algorithm, etc. The existing butler system has the following problems:

1. The algorithm steps of the intelligent calculation appointment time and cost reminder function adopted in the existing system are complicated;

2. Excessive running time and running costs are required, and the results are not accurate enough;

3. It is difficult to effectively predict and provide reference for future behavior through data extraction of regular information, lacking scientific prediction ability.

Contents of the invention

In order to solve the above-mentioned problems existing in the prior art, the present invention is to design a cloud-based system with fast convergence speed, minimized running time and running cost, capable of real-time processing and prediction of data, and capable of providing users with high accuracy and high efficiency. Computing smart housekeeping system and method.

In order to achieve the above object, the technical scheme of the present invention is as follows:

A smart housekeeping system based on cloud computing, including enterprise-level clients and ordinary personal clients;

The enterprise-level client includes:

Product configuration module: used for product management and purchase agreement management, responsible for configuration management of different modules;

Product registration module: used to register products with different time limits and different levels. The product registration code is prefixed with the first letter of the pinyin of the three major categories, followed by a free code;

Product details management module: used to display the configured product details management and query according to conditions;

Product association configuration module: used to set a batch of other products associated with this product;

Portal management module: used for announcement management, information encyclopedia, focus map management, patent product relationship management and enterprise dynamic management;

Sales financial management module: used for customer service to check the progress of their orders; used for national financial managers to view all orders and financial data, and the data is divided according to the order area; used for sales payment managers to view WeChat payment and Alipay payment , Refund three types of logs;

Express administrator management module: for national express administrators to view all orders and express data; for regional express administrators to view the data in their region, and the data is divided according to the order area; for national order distributors Allocate all orders; it is used by the regional order allocator to allocate the orders in this region. The initial region of the order is taken from the user's region and can be changed by the administrator;

The common personal client includes:

Settlement module: used to remind users in time and provide self-settlement functions;

User information module: used to improve user information and recipient information;

Butler service module: used to view purchased products, query data reports, upload data, download data;

Data report query module: used to summarize and query the user's income, purchase, and expenditure data;

Data navigation module: used to provide a navigation path guidance function for quickly obtaining relevant data.

A working method of a smart housekeeping system based on cloud computing, comprising the following steps:

A. Establish a unified mathematical model for the efficiency and stability of the smart housekeeper system;

The design of the unified mathematical model of stable efficiency and stability is based on the introduction of double-chain cloud quantum methods to generate "event chain" and "time chain" populations. The smaller the value, the higher the rescheduling efficiency and stability, and the design efficiency And the unified mathematical model of stability is as follows:

In the formula: F _n represents the management completion time of the event n to be managed; b _n represents the time to respond to the event n to be managed; _{DL n represents} the completion time window of the event n to be managed; In the scheduling and initial scheduling, the module m of the smart housekeeper system starts to manage the event n to be managed; PF() represents the deviation penalty value function; RT represents the current management time; n _i represents the total number of events that the user needs to complete at the current moment; m represents the serial number of the module, that is, each module in the system is represented by the natural serial number in turn;

B. Update the data stored in the smart housekeeper system in real time, and perform data cleaning and cluster analysis processing according to the data size and type. In the sales finance module, settlement module and data report query module, in order to obtain data cleaning and data The optimal solution of cluster analysis, improve the convergence speed of the obtained solution, and optimize the cost variable. The steps to obtain the crossover operator based on the cloud adaptive genetic algorithm are as follows:

B1. Denote the two individuals participating in the crossover operation as the parent I ^F and the mother I ^M respectively;

B2. Select two random integers q ₁ and q ₂ from the range [1, 10], and generate two sub-individuals on this basis;

B3. Generate ID ^. For the product list λ ^D of ID, the first q ¹ positions are determined by the _first q ₁ positions of the parent individual I ^M , and other positions are determined by the λ ^F of the parent individual;

B4. For the service resource allocation list ^{N D of} ID, its first q ₂ positions are determined by the first q ₂ positions of the parent individual I ^M , and the other positions are determined by the ^NF of the parent individual; similarly, the interaction between the parent and the mother Generate another sub-individual I ^T ;

B5. For the list of products, the exchange type and the insertion type are adopted: the exchange type refers to the exchange of different products in the product list, if the relationship constraints are not satisfied after the exchange, then the original position is replaced, and the next exchange is entered; the insertion type refers to the first Calculate the last position lc ₁ of all predecessor products in the list of the mutation operator and the front position lc ₂ of all subsequent products in the list, and then randomly select the position lc in lc ₁ and lc ₂ , and insert this operator into lc Insertion mutation is a more mutation method. The individual mutation step is to firstly mutate each product j _i of the product list λ in individual I according to the probability, and perform operator position transformation according to the insertion mutation method; Then assign each administrator of the service resource list in individual I For the product corresponding to this location, randomly allocate services that meet the demand from N _i ; generate cloud crossover probability and mutation probability through the Y conditional cloud generator and forward cloud generator, and generate crossover operators and mutation operators through the X conditional cloud generator ; λ is the event list that satisfies the sequence constraints of all management tasks; N is the allocation list of the management module, which represents the vector composed of the distribution mode corresponding to each event in the event list, i represents the serial number of the managed event, l _c represents the operator insertion s position;

C. Feedback and store the processing results through analysis and provide customers with predictable reference information: predictable reference information includes expenditure flow, purchase and consumption interest point prediction, and payment category tendency, and realizes the selection of the optimal search solution , the settlement module and the courier manager’s response module obtain the non-dominated optimal solution through the cloud computing mentioned above. For the optimal solution obtained in these two modules, the hierarchical method of AHP is adopted, that is, from top to bottom, it includes the total target layer, sub-targets Layer and program layer; the overall goal layer is the overall goal of solving real-time management problems, the sub-goal layer includes the sub-goals of minimizing total cost, minimizing running time and maximizing customer satisfaction, and the solution layer is based on a set of non- The design of the dominant solution set; the steps of solving the optimal solution of the "smart housekeeper system" based on the cloud adaptive genetic algorithm based on the AHP strategy are as follows, and each individual corresponds to a set of solutions of the above-mentioned settlement module and express manager response module:

C1. Randomly generate x individuals to form the initial population, and initialize p _cr and p _ct ; p _cr represents the cloud crossover operator, and p _ct represents the cloud mutation operator;

C2. Solve the fitness of each individual, and adjust p _cr and p _ct adaptively;

C3. Use the optimal retention strategy to perform the selection operation;

C4. Search for the fitness value and judge whether there is an individual update. If there is no update within 30 consecutive generations, the population number will be doubled;

C5. Judging whether the optimal solution has been obtained, if obtained, then output the best individual, otherwise perform step C2;

D. After the user's data management and tendency judgment tend to be stable, the data management and prediction functions are completed. In the settlement module and the courier administrator response module, in order to optimize the real-time prediction response time of the smart housekeeper system, a system based on The cloud computing method obtains the optimal solution, and the specific steps are as follows:

D1. Initialize the population, and carry out the chromosome coding of the cloud genetic algorithm to represent the events to be responded in the queue; the population is the user's collection of events to be managed;

D2, the design expression is the fitness evaluation function of fit(x)=1/Z(x), Z(x) represents the individual objective function value, and the smaller the function value, the better the individual;

D3. Use the optimal individual retention strategy and fitness ratio selection to select individuals in the population to determine the range of individuals entering the next generation population;

D4. Use the X condition generator of the cloud model to generate the crossover operator p _cr , and perform the crossover operation on the parent individual; adopt the double crossover method, place the gene region between the intersection points at the top of the offspring individual, and remove the parent The same code in the individual, and copy other codes to the offspring in order; if the offspring individual exceeds the constraint condition, move the position 0 to adjust;

D5. Use the X condition generator of the cloud model to generate the mutation operator p _ct , and perform an exchange mutation operation on the individual; randomly select two codes r ₁ and r ₂ from the genetic codes of the mutant individual, and r ₁ and r ₂ are Non-zero natural numbers, exchange the selected codes to generate new individuals;

D6. Rebuild a new population by selecting k excellent individuals from the parent population and k individuals from the offspring population, thereby expanding the population size and increasing the search space; through the related operations of the genetic algorithm, re-obtain k next-generation populations; This step of operation can better preserve the excellent genes of the population, and obtain better feasible solutions and optimal solutions;

D7, updating the quantum gate;

D8. Judging whether the stop condition is satisfied, if not, skip to step D3, otherwise stop the algorithm.

Compared with the prior art, the present invention has the following beneficial effects:

1. Since the present invention adopts the basic cloud self-adaptive genetic algorithm crossover operator, it realizes the minimization of the operating cost of the butler management system and the timely and accurate processing of uncertain information.

2. Since the present invention adopts the cloud-adaptive genetic algorithm based on the AHP strategy to solve the optimal solution strategy of the "smart housekeeper system", the optimal decision-making scheme can be obtained from a group of obtained non-dominated solutions.

3. Due to the data cleaning, denoising and clustering analysis algorithms designed by the present invention, the prediction of the user's future behavior is realized and an optimal butler system management reference solution is provided.

4. The method based on cloud computing proposed by the present invention improves the speed of optimal solution convergence, increases the search range of the global optimal solution, reduces the running time and cost of the smart housekeeper system, and improves the operating efficiency and shortening of the system. response time.

5. To sum up, the present invention designs a cloud computing-based algorithm to solve the problems in the existing system, such as redundant algorithm steps, excessive time and operating costs, inaccurate results, and lack of predictive functions. The smart housekeeper system and method can realize the fast convergence speed of obtaining the optimal solution of a reasonable scheduling plan in the housekeeper system; minimize the running time of the algorithm and reduce the running cost; use cloud computing methods to realize data cleaning and cluster analysis, and realize real-time Predictive processing function; it can provide users with a high-accuracy and high-efficiency housekeeping system management and scheduling plan.

Description of drawings

Fig. 1 is a schematic diagram of the system structure of the present invention.

Fig. 2 is a flow chart of the method of the present invention.

detailed description

The present invention will be further described below in conjunction with the accompanying drawings. A smart housekeeper system based on cloud computing, including enterprise-level client and ordinary personal client; the specific composition is shown in Figure 1. A working method of a smart housekeeping system based on cloud computing. The specific steps are shown in Figure 2. The module serial number m in step A is compiled as follows: for example, 1 represents the product configuration module, and 2 represents the product registration module; until all natural numbers are used to represent all modules.

The method of the present invention can be implemented by an embedded chip, a software module executed by a processor, or a combination of the two. Software modules can be placed in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other Any other known storage medium.

The specific embodiments described above have further described the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above descriptions are only specific embodiments of the present invention and are not intended to limit the protection of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims (1)

1. A smart housekeeping system based on cloud computing, including enterprise-level client terminals and ordinary personal client terminals; It is characterized in that: the enterprise-level client includes: Product configuration module: used for product management and purchase agreement management, responsible for configuration management of different modules; Product registration module: used to register products with different time limits and different levels. The product registration code is prefixed with the first letter of the pinyin of the three major categories, followed by a free code; Product details management module: used to display the configured product details management and query according to conditions; Product association configuration module: used to set a batch of other products associated with this product; Portal management module: used for announcement management, information encyclopedia, focus map management, patent product relationship management and enterprise dynamic management; Sales financial management module: used for customer service to check the progress of their orders; used for national financial managers to view all orders and financial data, and the data is divided according to the order area; used for sales payment managers to view WeChat payment and Alipay payment , Refund three types of logs; Express administrator management module: for national express administrators to view all orders and express data; for regional express administrators to view the data in their region, and the data is divided according to the order area; for national order distributors Allocate all orders; it is used by the regional order allocator to allocate the orders in this region. The initial region of the order is taken from the user's region and can be changed by the administrator; The common personal client includes: Settlement module: used to remind users in time and provide self-settlement functions; User information module: used to improve user information and recipient information; Butler service module: used to view purchased products, query data reports, upload data, download data; Data report query module: used to summarize and query the user's income, purchase, and expenditure data; Data navigation module: used to provide a navigation path guidance function for quickly obtaining relevant data. A working method of a smart housekeeping system based on cloud computing, characterized in that: comprising the following steps: A. Establish a unified mathematical model for the efficiency and stability of the smart housekeeper system; The design of the unified mathematical model of stable efficiency and stability is based on the introduction of double-chain cloud quantum methods to generate "event chain" and "time chain" populations. The smaller the value, the higher the rescheduling efficiency and stability, and the design efficiency And the unified mathematical model of stability is as follows: <mrow> <mi>min</mi> <mi> </mi> <mi>Z</mi> <mo>=</mo> <mn>5</mn> <mo>&amp;CenterDot;</mo> <mrow> <mo>(</mo> <mrow> <mi>max</mi> <mrow> <mo>(</mo> <msub> <mi>F</mi> <mi>n</mi> </msub> <mo>)</mo> </mrow> <mo>-</mo> <mi>min</mi> <mrow> <mo>(</mo> <msub> <mi>b</mi> <mi>n</mi> </msub> <mo>)</mo> </mrow> </mrow> <mo>)</mo> </mrow> <mo>+</mo> <mn>2</mn> <mo>&amp;CenterDot;</mo> <msub> <mi>&amp;Sigma;&amp;psi;</mi> <mi>n</mi> </msub> <mrow> <mo>(</mo> <mrow> <msub> <mi>F</mi> <mi>n</mi> </msub> <mo>-</mo> <msub> <mi>DL</mi> <mi>n</mi> </msub> </mrow> <mo>)</mo> </mrow> <mo>+</mo> <mn>4</mn> <mo>&amp;CenterDot;</mo> <mfrac> <mrow> <mo>&amp;lsqb;</mo> <mrow> <mi>&amp;Sigma;</mi> <mi>&amp;Sigma;</mi> <mo>|</mo> <msub> <msup> <mi>t</mi> <mo>&amp;prime;</mo> </msup> <mrow> <mi>n</mi> <mi>m</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>t</mi> <mrow> <mi>n</mi> <mi>m</mi> </mrow> </msub> <mo>|</mo> <mo>+</mo> <mi>&amp;Sigma;</mi> <mi>&amp;Sigma;</mi> <mi>P</mi> <mi>F</mi> <mrow> <mo>(</mo> <mrow> <msub> <mi>t</mi> <mrow> <mi>n</mi> <mi>m</mi> </mrow> </msub> <mo>+</mo> <msub> <msup> <mi>t</mi> <mo>&amp;prime;</mo> </msup> <mrow> <mi>n</mi> <mi>m</mi> </mrow> </msub> <mo>-</mo> <mn>2</mn> <mi>R</mi> <mi>T</mi> </mrow> <mo>)</mo> </mrow> </mrow> <mo>&amp;rsqb;</mo> </mrow> <mrow> <munderover> <mi>&amp;Sigma;</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>n</mi> <mi>i</mi> </msub> <mo>/</mo> <mi>n</mi> </mrow> </mfrac> </mrow> In the formula: F _n represents the management completion time of the event n to be managed; b _n represents the time to respond to the event n to be managed; _{DL n represents} the completion time window of the event n to be managed; In the scheduling and initial scheduling, the module m of the smart housekeeper system starts to manage the event n to be managed; PF() represents the deviation penalty value function; RT represents the current management time; n _i represents the total number of events that the user needs to complete at the current moment; m represents the serial number of the module, that is, each module in the system is represented by the natural serial number in turn; B. Update the data stored in the smart housekeeper system in real time, and perform data cleaning and cluster analysis processing according to the data size and type. In the sales finance module, settlement module and data report query module, in order to obtain data cleaning and data The optimal solution of cluster analysis, improve the convergence speed of the obtained solution, and optimize the cost variable. The steps to obtain the crossover operator based on the cloud adaptive genetic algorithm are as follows: B1. Denote the two individuals participating in the crossover operation as the parent I ^F and the mother I ^M respectively; B2. Select two random integers q ₁ and q ₂ from the range [1, 10], and generate two sub-individuals on this basis; B3. Generate ID ^. For the product list λ ^D of ID, the first q ¹ positions are determined by the _first q ₁ positions of the parent individual I ^M , and other positions are determined by the λ ^F of the parent individual; B4. For the service resource allocation list ^{N D of} ID, its first q ₂ positions are determined by the first q ₂ positions of the parent individual I ^M , and the other positions are determined by the ^NF of the parent individual; similarly, the interaction between the parent and the mother Generate another sub-individual I ^T ; B5. For the list of products, the exchange type and the insertion type are adopted: the exchange type refers to the exchange of different products in the product list, if the relationship constraints are not satisfied after the exchange, then the original position is replaced, and the next exchange is entered; the insertion type refers to the first Calculate the last position lc ₁ of all predecessor products in the list of the mutation operator and the front position lc ₂ of all subsequent products in the list, and then randomly select the position lc in lc ₁ and lc ₂ , and insert this operator into lc Insertion mutation is a more mutation method. The individual mutation step is to firstly mutate each product j _i of the product list λ in individual I according to the probability, and perform operator position transformation according to the insertion mutation method; Then assign each administrator of the service resource list in individual I For the product corresponding to this location, randomly allocate services that meet the demand from N _i ; generate cloud crossover probability and mutation probability through the Y conditional cloud generator and forward cloud generator, and generate crossover operators and mutation operators through the X conditional cloud generator ; λ is the event list that satisfies the sequence constraints of all management tasks; N is the allocation list of the management module, which represents the vector composed of the distribution mode corresponding to each event in the event list, i represents the serial number of the managed event, l _c represents the operator insertion s position; C. Feedback and store the processing results through analysis and provide customers with predictable reference information: predictable reference information includes expenditure flow, purchase and consumption interest point prediction, and payment category tendency, and realizes the selection of the optimal search solution , the settlement module and the courier manager’s response module obtain the non-dominated optimal solution through the cloud computing mentioned above. For the optimal solution obtained in these two modules, the hierarchical method of AHP is adopted, that is, from top to bottom, it includes the total target layer, sub-targets Layer and program layer; the overall goal layer is the overall goal of solving real-time management problems, the sub-goal layer includes the sub-goals of minimizing total cost, minimizing running time and maximizing customer satisfaction, and the solution layer is based on a set of non- The design of the dominant solution set; the steps of solving the optimal solution of the "smart housekeeper system" based on the cloud adaptive genetic algorithm based on the AHP strategy are as follows, and each individual corresponds to a set of solutions of the above-mentioned settlement module and express manager response module: C1. Randomly generate x individuals to form the initial population, and initialize p _cr and p _ct ; p _cr represents the cloud crossover operator, and p _ct represents the cloud mutation operator; C2. Solve the fitness of each individual, and adjust p _cr and p _ct adaptively; C3. Use the optimal retention strategy to perform the selection operation; C4. Search for the fitness value and judge whether there is an individual update. If there is no update within 30 consecutive generations, the population number will be doubled; C5. Judging whether the optimal solution has been obtained, if obtained, then output the best individual, otherwise perform step C2; D. After the user's data management and tendency judgment tend to be stable, the data management and prediction functions are completed. In the settlement module and the courier administrator response module, in order to optimize the real-time prediction response time of the smart housekeeper system, a system based on The cloud computing method obtains the optimal solution, and the specific steps are as follows: D1. Initialize the population, and carry out the chromosome coding of the cloud genetic algorithm to represent the events to be responded in the queue; the population is the user's collection of events to be managed; D2, the design expression is the fitness evaluation function of fit(x)=1/Z(x), Z(x) represents the individual objective function value, and the smaller the function value, the better the individual; D3. Use the optimal individual retention strategy and fitness ratio selection to select individuals in the population to determine the range of individuals entering the next generation population; D4. Use the X condition generator of the cloud model to generate the crossover operator p _cr , and perform the crossover operation on the parent individual; adopt the double crossover method, place the gene region between the intersection points at the top of the offspring individual, and remove the parent The same code in the individual, and copy other codes to the offspring in order; if the offspring individual exceeds the constraint condition, move the position 0 to adjust; D5. Use the X condition generator of the cloud model to generate the mutation operator p _ct , and perform an exchange mutation operation on the individual; randomly select two codes r ₁ and r ₂ from the genetic codes of the mutant individual, and r ₁ and r ₂ are Non-zero natural numbers, exchange the selected codes to generate new individuals; D6. Rebuild a new population by selecting k excellent individuals from the parent population and k individuals from the offspring population, thereby expanding the population size and increasing the search space; through the related operations of the genetic algorithm, re-obtain k next-generation populations; This step of operation can better preserve the excellent genes of the population, and obtain better feasible solutions and optimal solutions; D7, updating the quantum gate; D8. Judging whether the stop condition is satisfied, if not, skip to step D3, otherwise stop the algorithm.