CN117131434A - Intelligent prediction equipment suitable for people flow prediction and reasonable regulation and control of school - Google Patents
Intelligent prediction equipment suitable for people flow prediction and reasonable regulation and control of school Download PDFInfo
- Publication number
- CN117131434A CN117131434A CN202310970411.2A CN202310970411A CN117131434A CN 117131434 A CN117131434 A CN 117131434A CN 202310970411 A CN202310970411 A CN 202310970411A CN 117131434 A CN117131434 A CN 117131434A
- Authority
- CN
- China
- Prior art keywords
- prediction
- module
- suburban
- people flow
- probability
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 241000282461 Canis lupus Species 0.000 claims abstract description 37
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 18
- 241000282421 Canidae Species 0.000 claims description 36
- 238000000034 method Methods 0.000 claims description 19
- 238000012544 monitoring process Methods 0.000 claims description 15
- 238000005070 sampling Methods 0.000 claims description 14
- 230000006855 networking Effects 0.000 claims description 10
- 238000000342 Monte Carlo simulation Methods 0.000 claims description 9
- 238000009826 distribution Methods 0.000 claims description 9
- 230000014509 gene expression Effects 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 6
- 238000005065 mining Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 6
- 230000001105 regulatory effect Effects 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 239000006185 dispersion Substances 0.000 claims description 3
- 238000005315 distribution function Methods 0.000 claims description 3
- 238000013508 migration Methods 0.000 claims description 3
- 230000005012 migration Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000009827 uniform distribution Methods 0.000 claims description 3
- 241001347978 Major minor Species 0.000 claims description 2
- 230000001186 cumulative effect Effects 0.000 claims 4
- 238000009825 accumulation Methods 0.000 description 8
- 230000002265 prevention Effects 0.000 description 5
- 230000001276 controlling effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012614 Monte-Carlo sampling Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000005802 health problem Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 150000007523 nucleic acids Chemical class 0.000 description 1
- 102000039446 nucleic acids Human genes 0.000 description 1
- 108020004707 nucleic acids Proteins 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses intelligent prediction equipment suitable for people flow prediction and reasonable regulation of schools; the system comprises a data acquisition unit, a data processing and storage module, a people flow prediction unit, a people flow probability prediction module, a flow regulation and control unit, an access control module, an administrator scheduling module and a educational administration network module which are connected with each other. The VMD-KELM people flow prediction model provided by the invention relates to factors such as weather factors, course and examination factors, holiday factors, historical people flow data and the like, and can be used as input of the people flow prediction model, so that the prediction accuracy is obviously improved; the VMD signal decomposition of the prediction model is optimized through a multi-target wolf algorithm, so that the accuracy of data is improved, the people flow and the probability of each area in the current day are reasonably predicted, and a regulation and control scheme is formulated; in addition, the invention predicts the people flow and probability of each area through the prediction model, so that the related expenditure of schools is obviously reduced.
Description
Technical Field
The invention relates to the technical field of intelligent algorithms and prediction, in particular to intelligent prediction equipment suitable for people flow prediction and reasonable regulation of schools.
Background
As the level of diathesis education increases, schools are increasingly large in scale, and the traffic and density between school sub-areas gradually tend to saturate. Schools are important points for epidemic prevention as places with high people flow and people concentration. The increase in the flow rate and density of people brings about the best management cost and epidemic prevention cost, and therefore, the purposes of reducing the cost and ensuring the safety are achieved. Schools must appreciate cost control. In the regional people stream management process, there are also many problems, such as randomness of the people stream in partial regions, for the same region, large-scale people gathering exists under specific conditions, and the situation of low flow or no people exists for several consecutive days. The expenditure of school on people stream management mainly comes from equipment use, electricity consumption and epidemic prevention expenditure, a specific branch management and control plan is not formulated, and only excessive manpower and material resources can be input for management and control in the face of random unknown people stream. This has a certain impediment to the cost of controlling school people stream management expenditures, increasing school safety.
Therefore, aiming at the existing problems of the schools at the current stage, equipment is needed to provide a people flow regulation scheme in real time so as to solve the problem that people gather at the randomness of people flow, regulate and control the people flow in each area through reasonable current limiting and course time regulation, help the schools to effectively control the cost, ensure the health and safety of teachers and students at the schools while compressing the cost rate, and achieve the aims of economy and safety.
Disclosure of Invention
Aiming at the problems, the invention provides intelligent prediction equipment suitable for people flow prediction and reasonable regulation of schools, which can predict people flow in each time period of an area, regulate course time through people flow prediction data, and provide a reasonable course in a current limiting scheme so as to solve unsafe and health problems caused by dense people flow of the current schools.
The technical scheme of the invention is as follows: the invention relates to intelligent prediction equipment suitable for people flow prediction and reasonable regulation of schools, which comprises a data acquisition unit, a data processing and storage module, a people flow prediction unit, a people flow probability prediction module, a people flow regulation unit, an access control module, an administrator scheduling module and a educational administration network module which are connected with each other;
the output end of the data acquisition unit is connected to the input end of the data processing and storage module, the output end of the data processing and storage module is connected to the input end of the traffic prediction unit, the output end of the traffic prediction unit is connected to the input end of the traffic probability prediction module, and the output end of the traffic probability prediction module is connected to the input end of the traffic regulation unit.
Further, the data acquisition unit comprises a weather factor acquisition module (weather factor), a course and examination factor acquisition module (course and examination factor), a holiday factor acquisition module (holiday factor) and a historical people flow data acquisition module (historical people flow data);
the weather factor acquisition module, the course and examination factor acquisition module, the holiday factor acquisition module and the historical traffic data acquisition module are all electrically connected to the corresponding monitoring switchboard.
Further, the data processing and storing module processes and stores all acquired data;
the people number is predicted by the people flow prediction unit by using a VMD-KELM people flow prediction model based on an improved suburban wolf algorithm;
the people stream probability prediction unit uses Gaussian distribution fitting to predict an error sequence to obtain an error accumulation probability density distribution function, then uses a Monte Carlo method to randomly sample the accumulation probability density for a plurality of times, obtains a corresponding error sequence according to the sampling accumulation probability density, and reorders the sampling error sequence according to the size of the sampling accumulation probability density; and finally, giving confidence coefficient, and combining the point prediction results to obtain a people stream probability interval corresponding to the corresponding moment.
Further, the people flow regulation and control unit comprises a control module, a networking module and a monitoring module;
the control module, the networking module and the monitoring module are all electrically connected to the monitoring switchboard.
Further, the total output end of the people stream regulating and controlling unit is divided into three thin output ends which are respectively connected with the access control module, the manager scheduling module and the educational administration network module; the output end of the educational administration network module is connected to the data processing and storing module;
the access control module, the manager scheduling module and the educational administration network module are all electrically connected to the monitoring switchboard.
Further, the optimization method of the intelligent prediction equipment suitable for the people flow prediction and reasonable regulation of the school is characterized in that data are collected through a data acquisition unit to establish a VMD-KELM people flow prediction model of the people flow prediction unit;
the VMD-KELM people flow prediction model is specifically realized as follows:
(1): setting and selecting RBF radial basis functions, and determining nodes of an input layer and an output layer;
(2): establishing a VMD algorithm model to perform signal decomposition on the collected data;
(3): the accuracy of the number K of sub-modes of the VMD algorithm model and the punishment factor alpha parameter is improved through finding the major-minor signal decomposition, 80% of all data are used as training samples, the other 20% are used as test sample training models, and a probability interval prediction model is used for predicting a people stream probability interval;
(4): inputting related parameters of weather factors, courses and examination factors and holiday factors, predicting, and adjusting the courses and examination factors through a educational administration network to find out optimal courses and examination arrangement of people's flows;
and the entrance guard module and the personnel management personnel scheduling module limit related personnel flows and schedule the personnel through the personnel flow control module.
Further, in the step (1), the setting selects an RBF radial basis function, and the implementation process of determining the input layer node and the output layer node is as follows:
(1.1): setting and selecting RBF radial basis functions:
wherein mu t As a center point of the lens, the lens is,is the radial base width; the radial basis width determines the speed of the radial basis function falling;
determining 4 nodes of an input layer, x 1 、x 2 、x 3 、x 4 Weather factors, course and examination factors, holiday factors, historical people flow data at the same time;
determining 1 node of an output layer, wherein Y represents a current-day people flow prediction result;
(1.2): establishing a VMD algorithm model to perform signal decomposition on the collected data; the algorithm formula is as follows:
wherein: u (u) k ＝{u 1 ,u 2 ,...,u k And omega k ＝{ω 1 ,ω 2 ,...,ω k K sub-mode sets decomposed by VMD and corresponding center frequency sets;to derive an operator; * Is a convolution operator; delta is a unit pulse function; j denotes that the imaginary input signal v (t) is an input signal;
normalizing the data processed by the data processing and storing module, and mapping the original value to the value x' of the interval [0,1] through the maximum-minimum normalization, wherein the mapping formula is as follows:
wherein max A and min A respectively represent the maximum value and the minimum value of the factor A;
randomly selecting 80% of all data as training samples and the rest 20% as test samples;
(1.3): and optimizing the number K of sub-modes and penalty factors alpha of the decomposition of the variation modal decomposition algorithm by adopting an improved suburban wolf algorithm.
Further, in the step (1.3), the specific steps of optimizing the number of sub-modes K and the penalty factor α of the decomposition of the variation modal decomposition algorithm by adopting the improved suburban wolf algorithm are as follows:
(1.3.1): inputting the number K of sub-modes and punishment factor alpha parameters to generate suburban wolf population X, and randomly initializing X c,j And set up the population N, group number N P Number N of suburban wolves of each group c Maximum number of evaluation times N of minimum objective function of prediction error NEF The method comprises the steps of carrying out a first treatment on the surface of the The initialization formula is as follows:
X c,j ＝l j +rand×(u j -l j )；
wherein: x is X c,j Social state factor of the c suburban wolf (j) th dimension, u j And l j Respectively represent the upper limit and the lower limit of the j-th dimension, j E [1, D]D represents a problem dimension; rand represents a uniform distribution in [0,1]]Random numbers of (a);
(1.3.2): calculating a fitness value, determining a global optimal suburban wolf, and setting the current iteration number to 0;
(1.3.3): current iteration number t<And when T, performing a dynamic grouping strategy; the algorithm emphasizes the exploratory capacity earlier, if N in the case of a fixed population N P Larger, the more packets, the N c The method is small, the stronger global optimal suburban wolf is obtained, and the local searching capability is enhanced; post-emphasis on capacity, if N P Smaller, then N c The mining speed of the global optimal suburban wolves is increased due to the fact that the mining speed is high;
(1.3.4): the growth of the optimal suburban wolves in the group is promoted by a global optimal suburban wolf guiding strategy, and the global optimal suburban wolf guiding strategy has the following formula:
the early stage of the algorithm: adopting Levy to fly the overall optimal suburban wolves to guide the optimal suburban wolves in the group;
step＝k/|m| 1/beta
new-alpha＝alpha+stepsize；
wherein: rand is a random number, k, m and randn are random vectors subject to normal distribution; alpha represents the optimal suburban wolves in the group; beta is a parameter, and the value is 1.5; x is X g Representing globally optimal suburban wolves;
and (3) at the later stage of the algorithm:
new-alpha＝alpha+(1-(1-nef/N NEF ))(X g -alpha)+(1-nef/N NEF )(X 1 -X 2 )；
wherein: nef is the current function evaluation number; x is X 1 And X 2 Representing randomly selected suburban wolves; (X) g -alpha) and (X) 1 -X 2 ) The sum of the two weights is 1;
(1.3.5): the enhancement of the overall performance of the population through the enhancement strategy of the worst suburban wolves requires enhancement of the social adaptability of the worst individuals, and the enhancement of the worst suburban wolves is divided into two stages of early stage and later stage, and the expressions are as follows:
the early expression:
wherein: x is X w Representing the worst suburban wolves within the group; rand is uniformly distributed in [0,1] for each component]Is a random number vector of (a);
the late expression:
new_w＝X w +(0.5+0.5×rand)(X g -X w +alpha-X w )；
(1.3.6): the individual information use degree in the group is improved by utilizing the dynamically adjusted information sharing growth strategy, and the formula is as follows:
new_X i ＝X i +(1-(1-nef/N NEF ))S 1 +(1-nef/N NEF )r 2 S 2
S 1 ＝r 1 (alpha-X 1 )+(1-r 1 )(cult-X 2 )
S 2 ＝(X 1 -X 2 )；
wherein: r is (r) 1 And r 2 Two random weights respectively; s is S 1 Is an optimized growth formula; s is S 2 Representing intra-group information sharing;
(1.3.7): heuristic cross migration strategy:
v＝ceil((i-1)×rand)
c r ＝(sin(2πt/4+π)t/T+1)/2；
wherein: i represents an index of the immigrating habitat; v represents an index of the selected immigrating habitat in the sample;
ceil () represents a round-up function; u (U) i,j (t+1) represents a new solution generated by the crossover; c r Representing the crossover probability;
(1.3.8): parallel control boundary, parallel calculation fitness value, updating global optimum suburban wolf;
(1.3.9): suburban wolf growing and death operations:
P s ＝1/D,P a ＝(1-P s )/2；
wherein: c 1 And c 2 Are randomly selected in the same groupSelecting suburban wolf indexes; j (j) 1 And j 2 Is two random dimensions of the newly-grown suburban wolves; r is R j Is the random number of the j-th dimension social state factor in the decision variable range; rand of j The value of the random number is [0,1] which is uniformly distributed]；P s Is the dispersion probability, P a The probability of the association is determined by the probability,
(1.3.10): judging whether the current iteration times t+1 are smaller than the maximum iteration times or not; repeating the steps (1.3.3) to (1.3.9); otherwise, go to step (1.3.11);
(1.3.11): outputting the optimal input sub-mode number K and penalty factor alpha parameters;
(1.3.12): predicting errors of data and a true value through a multi-time VMD-KELM human flow prediction model, and carrying out probability by using Gaussian function fitting through a human flow probability prediction module;
the people stream probability prediction module is estimated by using a Monte Carlo method; and randomly sampling the existing variables by the Monte Carlo method to obtain corresponding characteristic values of y, and calculating the probability distribution of y through a sampling result.
The beneficial effects of the invention are as follows: 1. the VMD-KELM people flow prediction model provided by the invention relates to a plurality of influencing factors, regional factors, weather factors, course and examination factors, holiday factors and historical people flow data. These factors play a key role in the people flow, and the key factors are used as the input of a people flow prediction model, so that the prediction accuracy can be obviously improved; 2. according to the method, VMD signal decomposition of the prediction model is optimized through the multi-target wolf algorithm, so that the accuracy of data is improved, and the data is prevented from being excessively decomposed. The people flow and the probability of each area of the day are reasonably predicted, and a regulation and control scheme is formulated, so that the waste of management epidemic prevention resources and profit and loss caused by excessive investment can be avoided; 3. according to the method, the people flow and the probability of each area are predicted through the prediction model, so that the phenomenon of excessive management investment is reduced, and the related expenditure of schools is obviously reduced.
Description of the drawings:
FIG. 1 is a structural frame diagram of the present invention;
FIG. 2 is a schematic diagram of the operation of the human flow prediction model of the present invention;
FIG. 3 is a flowchart illustrating the operation of the people stream probability prediction unit of the present invention;
FIG. 4 is a flow chart of a multi-strategy suburban wolf optimization algorithm adopted by the invention;
FIG. 5 is a graph showing the comparison of actual and predicted traffic for a day in an embodiment of the present invention;
FIG. 6 is a graph showing the comparison of the control expenditure before and after the use of the apparatus according to the embodiment of the present invention;
in the figure, 1 is a data acquisition unit, 11 is a weather factor acquisition module, 12 is a course and examination factor acquisition module, 13 is a holiday factor acquisition module, and 14 is a historical people flow data acquisition module;
2 is a data processing and storing module, 3 is a people flow prediction unit, and 4 is a people flow probability prediction module;
5 is a people flow regulation and control unit, 51 is a control module, 52 is a networking module, and 53 is a monitoring module;
and 6 is an access control module, 7 is an administrator scheduling module, and 8 is a educational administration network module.
Detailed Description
In order to more clearly describe the technical scheme of the invention, the technical scheme of the invention is further described in detail below with reference to the accompanying drawings:
as shown in fig. 1-4, the intelligent prediction device suitable for people flow prediction and reasonable regulation in schools comprises a data acquisition unit 1, a data processing and storage module 2, a people flow prediction unit 3, a people flow probability prediction module 4 and a people flow regulation unit 5;
the data acquisition unit 1 acquires data packet weather factors, course and examination factors, holiday factors and historical traffic data obtained from the weather factor acquisition module 11, course and examination factor acquisition module 12, holiday factor acquisition module 13 and historical traffic data acquisition module 14;
the data processing and storing module 2 processes and stores all acquired data;
the people flow prediction unit 3 predicts the number of people by using a VMD-KELM people flow prediction model based on an MS-COA algorithm;
the people flow probability prediction module 4 uses Gaussian distribution fitting to predict an error sequence to obtain an error accumulation probability density distribution function, then uses a Monte Carlo method to randomly sample the accumulation probability density for a plurality of times, obtains a corresponding error sequence according to the sampling accumulation probability density, and reorders the sampling error sequence according to the size of the sampling accumulation probability density; finally, giving confidence coefficient, and combining the point prediction results to obtain a people stream probability interval corresponding to the corresponding moment;
the people flow control unit 5 comprises control, networking, monitoring, access control, nucleic acid detection and educational administration net modules.
Further, the data acquisition unit 1 acquires data including weather factors, course and examination factors, holiday factors and historical people flow data, and inputs the data into the data processing and storing module 2; the data processing and storing module 2 stores a factor sequence related to the flow of people, a VMD decomposed flow sequence of people and the like.
Furthermore, the people flow prediction unit 3 predicts the people flow in different time periods of the day through a VMD-KELM prediction model (people flow prediction unit 3) by the data obtained by the data acquisition unit 1, and adopts an MS-COA algorithm to optimize model parameters;
furthermore, the people flow probability prediction module 4 uses a Monte Carlo construction probability interval prediction model to carry out probability prediction through a prediction result and an error obtained by a VMD-KELM prediction model (a people flow prediction unit 3), and gives the people flow probability at each time and the upper and lower limits of a confidence interval thereof through a Monte Carlo sampling method.
Further, the people flow regulating unit 5 comprises a control module 51, a networking module 52 and a monitoring module 53, the networking module 52 and the monitoring module 53 work cooperatively with the networking module, and the networking module 52 reads the data of the regional access control module 6 and the administrator regulating module 7, so that sudden situations of sudden increase of people flow are avoided;
the educational administration module 8 adjusts the next course and examination factor sequence and inputs the next course and examination factor sequence into the trained people stream and the probability prediction model thereof to predict, so that the maximum probability people stream in each time period of the area is ensured to be lower than a preset threshold value.
Further, an intelligent prediction device suitable for people flow prediction and reasonable regulation of schools is characterized in that data is collected through a data acquisition unit 1 of the intelligent prediction device to establish a VMD-KELM people flow prediction model, and the VMD-KELM people flow prediction model is realized as follows:
(1): setting and selecting RBF radial basis functions:
wherein mu t As a center point of the lens, the lens is,is the radial base width. The radial basis width determines the speed of the radial basis function falling;
(2): determining 4 nodes of an input layer, x 1 、x 2 、x 3 、x 4 Weather factors, course and examination factors, holiday factors, historical flow data at the same time;
(3): determining 1 node of an output layer, wherein Y represents a current-day people flow prediction result;
(4): and establishing a VMD algorithm model to perform signal decomposition on the collected data. The algorithm formula is as follows:
wherein: u (u) k ＝{u 1 ,u 2 ,...,u k And omega k ＝{ω 1 ,ω 2 ,...,ω k K sub-mode sets decomposed by VMD and corresponding center frequency sets;to derive an operator; * Is a convolution operator; delta is a unit pulse function; j denotes that the imaginary input signal v (t) is an input signal。
(5): normalizing the data processed by the data processing and storing module 2, and mapping the original value to the value x' of the interval [0,1] through the maximum-minimum normalization, wherein the mapping formula is as follows:
wherein max A and min A represent the maximum and minimum values of factor A, respectively;
(6): randomly selecting 80% of all data as training samples and the rest 20% as test samples;
(7): the specific steps of the number K of sub-modes and penalty factor alpha of the decomposition of the optimized Variation Modal Decomposition (VMD) algorithm by adopting the improved suburban wolf algorithm are as follows:
(8): inputting the number K of sub-modes and punishment factor alpha parameters to generate suburban wolf population X, and randomly initializing X c,j And set up the population N, group number N P Number N of suburban wolves of each group c Maximum number of evaluation times N of minimum objective function of prediction error NEF And the like. The initialization formula is as follows:
X c,j ＝l j +rand×(u j -l j )；
wherein: x is X c,j Social state factor of the c suburban wolf (j) th dimension, u j And l j Respectively represent the upper limit and the lower limit of the j-th dimension, j E [1, D]D represents a problem dimension; rand represents a uniform distribution in [0,1]]Random numbers of (a);
(9): calculating a fitness value, determining a global optimal suburban wolf, and setting the current iteration number to 0;
(10): current iteration number t<And when T, performing a dynamic grouping strategy; the algorithm emphasizes the exploratory capacity earlier, if N in the case of a fixed population N P Larger, the more packets, the N c The method is small, the stronger global optimal suburban wolf is obtained, and the local searching capability is enhanced; post-emphasis on capacity, if N P Smaller, then N c The mining speed of the global optimal suburban wolves is increased due to the fact that the mining speed is high;
(11): promoting growth of optimal suburban wolves in a group through a global optimal suburban wolf guiding strategy, and guiding the optimal suburban wolves globally
The policy formula is as follows:
the early stage of the algorithm: and adopting Levy to fly the globally optimal suburban wolves to guide the optimal suburban wolves in the group.
step＝k/|m| 1/beta
new-alpha＝alpha+stepsize；
Wherein: rand is a random number, k, m and randn are random vectors subject to normal distribution; alpha represents the optimal suburban wolves in the group; beta is a parameter, and the value is 1.5; x is X g Representing a globally optimal suburban wolf.
And (3) at the later stage of the algorithm:
new-alpha＝alpha+(1-(1-nef/N NEF ))(X g -alpha)+(1-nef/N NEF )(X 1 -X 2 )；
wherein: nef is the current function evaluation number; x is X 1 And X 2 Representing randomly selected suburban wolves. (X) g -alpha) and (X) 1 -X 2 ) The sum of the two weights is 1;
(12): the enhancement of the overall performance of the population through the enhancement strategy of the worst suburban wolves requires enhancement of the social adaptability of the worst individuals, and the enhancement of the worst suburban wolves is divided into two stages of early stage and later stage, and the expressions are as follows:
the early expression:
wherein: x is X w Representing the worst suburban wolves within the group; rand is uniformly distributed in [0,1] for each component]Is a random number vector of (a).
The late expression:
new_w＝X w +(0.5+0.5×rand)(X g -X w +alpha-X w )；
(13): the individual information use degree in the group is improved by utilizing the dynamically adjusted information sharing growth strategy, and the formula is as follows:
new_X i ＝X i +(1-(1-nef/N NEF ))S 1 +(1-nef/N NEF )r 2 S 2
S 1 ＝r 1 (alpha-X 1 )+(1-r 1 )(cult-X 2 )
S 2 ＝(X 1 -X 2 )；
wherein: r is (r) 1 And r 2 Two random weights respectively; s is S 1 Is an optimized growth formula; s is S 2 Representing intra-group information sharing;
(14): heuristic cross migration strategy:
v＝ceil((i-1)×rand)
c r ＝(sin(2πt/4+π)t/T+1)/2；
wherein: i represents an index of the immigrating habitat; v represents an index of the selected immigrating habitat in the sample; ceil () represents a round-up function; u (U) i,j (t+1) represents a new solution generated by the crossover; c r Representing the crossover probability;
(15): parallel control boundary, parallel calculation fitness value, updating global optimum suburban wolf;
(16): suburban wolf growing and death operations:
510)
P s ＝1/D,P a ＝(1-P s )/2；
wherein: c 1 And c 2 Two suburban wolf indexes randomly selected in the same group respectively; j (j) 1 And j 2 Is two random dimensions of the newly-grown suburban wolves; r is R j Is the random number of the j-th dimension social state factor in the decision variable range; rand of j The value of the random number is [0,1] which is uniformly distributed]；P s Is the dispersion probability, P a Associating probabilities;
(17): judging whether the current previous iteration times t+1 are smaller than the maximum iteration times or not; repeating the steps (10) to (16) if yes; otherwise, entering step (18);
(18): outputting the optimal input sub-mode number K and penalty factor alpha parameters;
(19): and predicting errors of data and a true value through a multi-time VMD-KELM human flow prediction model, and carrying out probability by using Gaussian function fitting through a probability interval prediction model, wherein the probability interval prediction module is estimated by using a Monte Carlo method. The Monte Carlo method randomly samples the variables to obtain corresponding characteristic values of y, and the probability distribution of y is calculated through a large number of sampling results.
As shown in fig. 5, the device has better prediction accuracy for people stream prediction, and people stream fluctuation in a monitoring area is larger during 8 points and 18 points of personnel fluctuation before regulation and control of the device, for example, people stream control is more difficult. However, after the device is regulated, the people flow in the area is obviously controlled, the predicted people flow fluctuation is higher, and after the device passes through, the people flow fluctuation in the area is smaller, so that the management is more convenient.
As shown in FIG. 6, the regulation scheme obtained by the prediction model provided by the invention takes 12 months of the last year as an example, compared with the previous year, the net expenditure is reduced by 31659 yuan, and the average net expenditure per month is reduced by 2638.25 yuan, so that the invention has good economical efficiency and practicability. Effectively solves the problems of excessive epidemic prevention expenditure and people flow management in schools.
Claims (10)
1. The intelligent prediction equipment suitable for people flow prediction and reasonable regulation of schools is characterized by comprising a data acquisition unit, a data processing and storage module, a people flow prediction unit, a people flow probability prediction module, a people flow regulation unit, an access control module, an administrator scheduling module and a educational administration network module which are connected with each other;
the output end of the data acquisition unit is connected to the input end of the data processing and storage module, the output end of the data processing and storage module is connected to the input end of the traffic prediction unit, the output end of the traffic prediction unit is connected to the input end of the traffic probability prediction module, and the output end of the traffic probability prediction module is connected to the input end of the traffic regulation unit.
2. The intelligent prediction device suitable for people flow prediction and reasonable regulation of schools according to claim 1, wherein the data acquisition unit comprises a weather factor acquisition module, a course and examination factor acquisition module, a holiday factor acquisition module and a historical people flow data acquisition module;
the weather factor acquisition module, the course and examination factor acquisition module, the holiday factor acquisition module and the historical traffic data acquisition module are all electrically connected to the corresponding monitoring switchboard.
3. The intelligent prediction device suitable for the prediction and reasonable regulation of the traffic of a person in a school according to claim 1, wherein the traffic prediction unit predicts the number of persons by using a VMD-KELM traffic prediction model based on a modified suburban wolf algorithm.
4. The intelligent prediction device suitable for people flow prediction and reasonable regulation and control of a school according to claim 1, wherein the people flow probability prediction unit uses Gaussian distribution to fit a prediction error sequence to obtain an error cumulative probability density distribution function, and then adopts a Monte Carlo method to randomly sample the cumulative probability density for a plurality of times;
obtaining a corresponding error sequence according to the sampling cumulative probability density, and reordering the sampling error sequence according to the size of the sampling cumulative probability density;
and finally, giving confidence coefficient, and combining the point prediction results to obtain a people stream probability interval corresponding to the corresponding moment.
5. The intelligent prediction device suitable for people flow prediction and reasonable regulation of schools according to claim 1, wherein the people flow regulation and control unit comprises a control module, a networking module and a monitoring module;
the control module, the networking module and the monitoring module are all electrically connected to the monitoring switchboard.
6. The intelligent prediction device for predicting and reasonably regulating the traffic of a person, which is applicable to schools, according to claim 1, wherein the total output end of the traffic regulation unit is divided into three thin output ends which are respectively connected with an access control module, an administrator scheduling module and a educational administration network module; the output end of the educational administration network module is connected to the data processing and storing module;
the access control module, the manager scheduling module and the educational administration network module are all electrically connected to the monitoring switchboard.
7. The optimization method of intelligent prediction equipment suitable for people flow prediction and reasonable regulation of schools according to any one of claims 1-6, wherein the data is collected by a data acquisition unit to establish a VMD-KELM people flow prediction model of the people flow prediction unit;
the implementation process of the VMD-KELM people flow prediction model is as follows:
(1): setting and selecting RBF radial basis functions, and determining nodes of an input layer and an output layer;
(2): establishing a VMD algorithm model to perform signal decomposition on the collected data;
(3): the accuracy of the number K of sub-modes of the VMD algorithm model and the punishment factor alpha parameter is improved through finding the major-minor signal decomposition, 80% of all data are used as training samples, the other 20% are used as test sample training models, and a probability interval prediction model is used for predicting a people stream probability interval;
(4): inputting parameters of weather factors, courses, examination factors and holiday factors of the alternate days, predicting, and adjusting the courses and examination factors through a educational administration network to find out optimal courses and examination arrangements of people's flows;
and the access control module and the manager scheduling module limit related people flows and schedule people through the people flow control unit.
8. The optimization method of intelligent prediction equipment suitable for people flow prediction and reasonable regulation in schools according to claim 7 is characterized in that,
in the step (1), the RBF radial basis function is set and used, and the implementation process of determining the nodes of the input layer and the output layer is as follows:
(1.1): setting and selecting RBF radial basis functions:
wherein mu t As a center point of the lens, the lens is,is the radial base width; the radial basis width determines the speed of the radial basis function falling;
determining 4 nodes of an input layer, x 1 、x 2 、x 3 、x 4 Weather factors, course and examination factors, holiday factors, historical people flow data at the same time;
determining 1 node of an output layer;
(1.2): establishing a VMD algorithm model to perform signal decomposition on the collected data; the algorithm formula is as follows:
wherein: u (u) k ＝{u 1 ,u 2 ,...,u k And omega k ＝{ω 1 ,ω 2 ,...,ω k K sub-mode sets decomposed by VMD and corresponding center frequency sets;to derive an operator; * Is a convolution operator; delta is a unit pulse function; j denotes that the imaginary input signal v (t) is an input signal;
normalizing the data processed by the data processing and storing module, and mapping the original value to the value x' of the interval [0,1] through the maximum-minimum normalization, wherein the mapping formula is as follows:
wherein max A and min A respectively represent the maximum value and the minimum value of the factor A;
randomly selecting 80% of all data as training samples and the rest 20% as test samples;
(1.3): and optimizing the number K of sub-modes and penalty factors alpha of the decomposition of the variation modal decomposition algorithm by adopting an improved suburban wolf algorithm.
9. The optimization method of intelligent prediction equipment suitable for people flow prediction and reasonable regulation and control in school according to claim 8, wherein in step (1.3), the specific steps of optimizing the number of sub-modes K and penalty factor α of the decomposition of the variation modal decomposition algorithm by adopting the improved suburban wolf algorithm are as follows:
(1.3.1): inputting the number K of sub-modes and punishment factor alpha parameters to generate suburban wolf population X, and randomly initializing X c,j And set up the population N, group number N P Number N of suburban wolves of each group c Maximum number of evaluation times N of minimum objective function of prediction error NEF The method comprises the steps of carrying out a first treatment on the surface of the The initialization formula is as follows:
X c,j ＝l j +rand×(u j -l j )；
wherein: x is X c,j Social state factor of the c suburban wolf (j) th dimension, u j And l j Respectively represent the upper limit and the lower limit of the j-th dimension, j E [1, D]D represents a problem dimension; rand represents a uniform distribution in [0,1]]Random numbers of (a);
(1.3.2): calculating a fitness value, determining a global optimal suburban wolf, and setting the current iteration number to 0;
(1.3.3): current iteration number t<And when T, performing a dynamic grouping strategy; the algorithm emphasizes the exploratory capacity earlier, if N in the case of a fixed population N P Larger, the more packets, the N c The method is small, the stronger global optimal suburban wolf is obtained, and the local searching capability is enhanced; post-emphasis on capacity, if N P Smaller, then N c The mining speed of the global optimal suburban wolves is increased due to the fact that the mining speed is high;
(1.3.4): the growth of the optimal suburban wolves in the group is promoted by a global optimal suburban wolf guiding strategy, and the global optimal suburban wolf guiding strategy has the following formula:
the early stage of the algorithm: adopting Levy to fly the overall optimal suburban wolves to guide the optimal suburban wolves in the group;
step＝k/|m| 1/beta
new-alpha＝alpha+stepsize；
wherein: rand is a random number, k, m and randn are random vectors subject to normal distribution; alpha represents the optimal suburban wolves in the group; beta is a parameter, and the value is 1.5; x is X g Representing globally optimal suburban wolves;
and (3) at the later stage of the algorithm:
new-alpha＝alpha+(1-(1-nef/N NEF ))(X g -alpha)+(1-nef/N NEF )(X 1 -X 2 )；
wherein: nef is the current function evaluation number; x is X 1 And X 2 Representing randomly selected suburban wolves; (X) g -alpha) and (X) 1 -X 2 ) The sum of the two weights is 1;
(1.3.5): the enhancement of the overall performance of the population through the enhancement strategy of the worst suburban wolves requires enhancement of the social adaptability of the worst individuals, and the enhancement of the worst suburban wolves is divided into two stages of early stage and later stage, and the expressions are as follows:
the early expression:
wherein: x is X w Representing the worst suburban wolves within the group; rand is uniformly distributed in [0,1] for each component]Is a random number vector of (a);
the late expression:
new_w＝X w +(0.5+0.5×rand)(X g -X w +alpha-X w )；
(1.3.6): the individual information use degree in the group is improved by utilizing the dynamically adjusted information sharing growth strategy, and the formula is as follows:
new_X i ＝X i +(1-(1-nef/N NEF ))S 1 +(1-nef/N NEF )r 2 S 2
S 1 ＝r 1 (alpha-X 1 )+(1-r 1 )(cult-X 2 )
S 2 ＝(X 1 -X 2 )；
wherein: r is (r) 1 And r 2 Two random weights respectively; s is S 1 Is an optimized growth formula; s is S 2 Representing intra-group information sharing;
(1.3.7): heuristic cross migration strategy:
v＝ceil((i-1)×rand)
c r ＝(sin(2πt/4+π)t/T+1)/2；
wherein: i represents an index of the immigrating habitat; v represents an index of the selected immigrating habitat in the sample; ceil () represents a round-up function; u (U) i,j (t+1) represents a new solution generated by the crossover; c r Representing the crossover probability;
(1.3.8): parallel control boundary, parallel calculation fitness value, updating global optimum suburban wolf;
(1.3.9): suburban wolf growing and death operations:
P s ＝1/D,P a ＝(1-P s )/2；
wherein: c 1 And c 2 Two suburban wolf indexes randomly selected in the same group respectively; j (j) 1 And j 2 Is two random dimensions of the newly-grown suburban wolves; r is R j Is the random number of the j-th dimension social state factor in the decision variable range; rand of j The value of the random number is [0,1] which is uniformly distributed]；P s Is the dispersion probability, P a The probability of the association is determined by the probability,
(1.3.10): judging whether the current iteration times t+1 are smaller than the maximum iteration times or not; repeating the steps (1.3.3) to (1.3.9); otherwise, go to step (1.3.11);
(1.3.11): outputting the optimal input sub-mode number K and penalty factor alpha parameters;
(1.3.12): and predicting errors of data and a true value through a multi-time VMD-KELM human flow prediction model, and carrying out probability by using Gaussian function fitting through a human flow probability prediction module.
10. The optimization method of intelligent prediction equipment suitable for people flow prediction and rational regulation in schools according to claim 9, wherein in step (1.3.12), the people flow probability prediction module is estimated by using the monte carlo method; and randomly sampling the existing variables by the Monte Carlo method to obtain corresponding characteristic values of y, and calculating the probability distribution of y through a sampling result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310970411.2A CN117131434B (en) | 2023-08-03 | Intelligent prediction equipment suitable for people flow prediction and reasonable regulation and control of school |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310970411.2A CN117131434B (en) | 2023-08-03 | Intelligent prediction equipment suitable for people flow prediction and reasonable regulation and control of school |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117131434A true CN117131434A (en) | 2023-11-28 |
| CN117131434B CN117131434B (en) | 2024-08-02 |
Family
ID=
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117830047A (en) * | 2024-01-15 | 2024-04-05 | 江苏科学梦创展科技有限公司 | Operation pressure prediction and intelligent adjustment system for science and technology exhibition hall |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113361777A (en) * | 2021-06-08 | 2021-09-07 | 淮阴工学院 | Runoff prediction method and system based on VMD decomposition and IHHO optimization LSTM |
| KR102357236B1 (en) * | 2021-06-25 | 2022-02-08 | 주식회사 빅스터 | Availability index prediction system according to the floating population by region and the confirmed population of infectious diseases |
| CN115829097A (en) * | 2022-11-22 | 2023-03-21 | 苏州科技大学 | Air conditioner ultra-short term load prediction method based on VMD and KELM |
| CN115994785A (en) * | 2023-01-09 | 2023-04-21 | 淮阴工学院 | Intelligent prediction method and system for catering traffic stock |
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113361777A (en) * | 2021-06-08 | 2021-09-07 | 淮阴工学院 | Runoff prediction method and system based on VMD decomposition and IHHO optimization LSTM |
| KR102357236B1 (en) * | 2021-06-25 | 2022-02-08 | 주식회사 빅스터 | Availability index prediction system according to the floating population by region and the confirmed population of infectious diseases |
| CN115829097A (en) * | 2022-11-22 | 2023-03-21 | 苏州科技大学 | Air conditioner ultra-short term load prediction method based on VMD and KELM |
| CN115994785A (en) * | 2023-01-09 | 2023-04-21 | 淮阴工学院 | Intelligent prediction method and system for catering traffic stock |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117830047A (en) * | 2024-01-15 | 2024-04-05 | 江苏科学梦创展科技有限公司 | Operation pressure prediction and intelligent adjustment system for science and technology exhibition hall |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110705743B (en) | New energy consumption electric quantity prediction method based on long-term and short-term memory neural network | |
| CN106529732A (en) | Carbon emission efficiency prediction method based on neural network and random frontier analysis | |
| CN102395135A (en) | VLR (Visitor Location Register) user number predicting method based on gray system model neural network | |
| Bai et al. | Spectrum prediction based on improved-back-propagation neural networks | |
| CN109242136A (en) | A kind of micro-capacitance sensor wind power Chaos-Genetic-BP neural network prediction technique | |
| CN110472840A (en) | A kind of agricultural water conservancy dispatching method and system based on nerual network technique | |
| CN112363896A (en) | Log anomaly detection system | |
| CN105701280A (en) | Southern America white-leg shrimp pond culture water quality prediction method | |
| Slade | Simulation of selected genealogies | |
| CN111901134B (en) | Method and device for predicting network quality based on recurrent neural network model (RNN) | |
| CN105578472A (en) | Wireless sensor network performance online optimization and planning method based on immune theory | |
| CN117131434B (en) | Intelligent prediction equipment suitable for people flow prediction and reasonable regulation and control of school | |
| CN110298506A (en) | A kind of urban construction horizontal forecast system | |
| CN112860973B (en) | Tissue health assessment method based on social network analysis | |
| Chu et al. | Adaptive brainstorm optimisation with multiple strategies | |
| CN112052985B (en) | Middle-short-term low-voltage prediction method based on lightgbm | |
| CN117131434A (en) | Intelligent prediction equipment suitable for people flow prediction and reasonable regulation and control of school | |
| Bezerra et al. | A self-adaptive approach for particle swarm optimization applied to wind speed forecasting | |
| CN110991616B (en) | Method for predicting BOD of effluent based on pruning feedforward small-world neural network | |
| Zahoor et al. | Evolutionary computation technique for solving Riccati differential equation of arbitrary order | |
| CN114139783A (en) | Wind power short-term power prediction method and device based on nonlinear weighted combination | |
| CN113902492A (en) | Time-sharing electricity price prediction method and system | |
| Furze et al. | Mathematical methods to quantify and characterise the primary elements of trophic systems | |
| CN110288159A (en) | A kind of smart city construction Measurement Method | |
| CN113539511B (en) | Infectious disease transmission treatment model optimization method based on K nearest neighbor constraint optimization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |