CN109460925A - A kind of platform area group of assets Performance Evaluation Methods based on BP neural network - Google Patents

A kind of platform area group of assets Performance Evaluation Methods based on BP neural network Download PDF

Info

Publication number
CN109460925A
CN109460925A CN201811350487.0A CN201811350487A CN109460925A CN 109460925 A CN109460925 A CN 109460925A CN 201811350487 A CN201811350487 A CN 201811350487A CN 109460925 A CN109460925 A CN 109460925A
Authority
CN
China
Prior art keywords
performance
neural network
platform area
assets
layer
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.)
Pending
Application number
CN201811350487.0A
Other languages
Chinese (zh)
Inventor
任腾云
陈刚
王春波
周溢青
屈维意
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
State Grid Jiangsu Electric Power Co Ltd
Jiangsu Electric Power Information Technology Co Ltd
Original Assignee
State Grid Jiangsu Electric Power Co Ltd
Jiangsu Electric Power Information Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by State Grid Jiangsu Electric Power Co Ltd, Jiangsu Electric Power Information Technology Co Ltd filed Critical State Grid Jiangsu Electric Power Co Ltd
Priority to CN201811350487.0A priority Critical patent/CN109460925A/en
Publication of CN109460925A publication Critical patent/CN109460925A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • General Health & Medical Sciences (AREA)
  • Educational Administration (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Primary Health Care (AREA)
  • Water Supply & Treatment (AREA)
  • Public Health (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Complex Calculations (AREA)

Abstract

The platform area group of assets evaluation method based on BP neural network that the invention discloses a kind of, comprising: performance appraisal indices in platform area are standardized；BP neural network evaluation model is constructed, and performance value is normalized；According to the comprehensive performance of BP neural network model evaluation platform area group of assets.The present invention is by filtering out evaluation index relevant to platform area group of assets comprehensive performance, using the BP neural network study performance evaluating index based on LM algorithm optimization and the non-linear relation between platform area group of assets performance, and then BP neural network evaluation model is obtained, pass through the comprehensive performance of BP neural network model evaluation platform area group of assets.

Description

A kind of platform area group of assets Performance Evaluation Methods based on BP neural network
Technical field
The present invention relates to a kind of evaluation methods, and in particular to a kind of platform area group of assets performance appraisal based on BP neural network Method.
Background technique
With the rapid development of China's power industry, the unit-area management work of various regions power supply at present lacks effective management means And Performance Management.Unit-area management be to electric power enterprise in the management means of distribution network and power marketing a germplasm fly Jump, it requires electric power enterprise to have preferable basis and higher level and Xiang Shi in human, financial, and material resources and information management The marketing system answered.Unit-area management master be to solve some such as distribution lines are chaotic, line loss is excessively high, human factor is excessive, Arrearage is excessive, stealing is ferocious sticks up, lacks the problems such as supervision.It is platform area integrated pipe by constructing platform area group of assets comprehensive evaluation model The smooth development for managing real work provides scientific basis.
In reality, although various regions, office, city also carry out respective marketing system according to the pertinent regulations of provincial electric power company Necessary improvement, but be to continue to use original some classical management approach and means nearly all to go out after having tried one or two years mostly Such or such problem is showed, line loss is still excessively high, and practitioner is still difficult to manage and lack supervision etc., unit-area management Work is carried out very not smooth.
In recent years, the development of neural network theory and application are assessed for group of assets provides new approaches.Neural network assessment Group of assets can use the powerful self-learning ability of network, Generalization Ability and Nonlinear Processing ability without founding mathematical models It is fitted non-linear relation complicated between platform area performance and evaluation index, therefore using neural network algorithm can be with overall merit The performance of platform area group of assets.
Summary of the invention
In order to overcome the above-mentioned deficiencies of the prior art, the object of the present invention is to provide a kind of platform area based on BP neural network Group of assets Performance Evaluation Methods are calculated by filtering out evaluation index relevant to platform area group of assets comprehensive performance using based on LM Non-linear relation between the BP neural network study performance evaluating index and platform area group of assets performance of method optimization, and then obtain BP Neural network model of performance appraisal passes through the comprehensive performance of BP neural network model evaluation platform area group of assets.
In order to achieve the above-mentioned object of the invention, the present invention adopts the following technical scheme that:
A kind of platform area group of assets Performance Evaluation Methods based on BP neural network, it is characterised in that the method includes following Step:
Step 1: performance appraisal indices in platform area are standardized；
Step 2: building BP neural network evaluation model, and performance value is normalized；
Step 3: according to the comprehensive performance of BP neural network model evaluation platform area group of assets.
In the step 1, if platform area number is N, the three-level evaluation index in each area is M, the achievement of the area NGe Tai sample It imitates evaluation index and forms platform area performance feature vector, X, have:
Wherein, xijThe i-th row, jth column element for platform area performance feature vector, X, i=1,2 ..., N, j=1,2 ..., M；
Platform area performance evaluating index is standardized, is had:
Wherein, ZijFor xijAmount after standardization,For xijAverage value, sijFor xijVariance.
The step 2 the following steps are included:
Step 2-1: building BP neural network evaluation model:
BP neural network model includes input layer, hidden layer and output layer, and the transmission function f (a) between each layer is used Logsig function, has:
Wherein, the independent variable of a transmission function f (a) between each layer, 0 < f (a) < 1；
Step 2-2: platform area group of assets items performance evaluating index is normalized:
If platform area performance is d, platform area group of assets performance d is normalized using formula (6), is had:
Wherein, di' for the value after the normalization of i-th area's comprehensive performance, diFor the performance score in i-th area, dminFor The minimum value of all area's performance, dmaxFor the maximum value in all area's performance, α, β are constant, and 0.9 < α <, 1,0 < β < 0.1。
The step 3 the following steps are included:
Step 3-1: using BP neural network model to ZijAnd di' carry out learning training；
Step 3-2: platform area performance evaluating index is brought into BP neural network model, calculates platform area comprehensive performance d.
In the step 3-1, for any area, since the performance evaluating index in each area is M, so input layer Containing M BP neuron, input layer input vector is set then as Zr=(Z1,Z2,…,Zm,…,ZM)T, the output vector of hidden layer For Yr=(y1,y2,…,yp,…,yP)T, the output vector of output layer is or=(o1,o2,…,ol,…,oL)T, desired output to Amount is dr=(d1,d2,…,dl,…,dL)T, wherein T indicates transposition, ZmFor m-th of BP neuron of input layer, ypFor hidden layer P-th of BP neuron, o1For the 1st neuron of output layer, d1For the 1st desired throughput, P is the BP nerve of hidden layer First number, L are the BP neuron number of output layer, m=1,2 ..., M, p=1,2 ..., P, l=1,2 ..., L；Input layer is to hidden Weight and threshold value containing layer are respectively wmpAnd bmp, the weight and threshold value of hidden layer to output layer are respectively wplAnd bpl；
Then, using BP neural network model to ZijAnd di' carry out learning training forward-propagating process it is as follows:
Output error e is indicated are as follows:
Using BP neural network model to ZijAnd di' carry out learning training back-propagation process it is as follows:
In (n+1)th iterative process, e is unfolded by Taylor's formula, obtains formula (10):
E (w (n+1))=e (w (n))+gT(n)Δw(n)+0.5ΔwT(n)A(n)Δw(n) (10)
Wherein, w (n) is the weight during nth iteration, and e (w (n)) is output error during nth iteration；w(n It+1) is the weight in (n+1)th iterative process, e (w (n+1)) is the output error in (n+1)th iterative process；G (n) is ladder Vector is spent, T indicates transposition；Δ w (n) is the variable quantity of (n+1)th with weight during nth iteration, i.e. Δ w (n)=w (n+1)- W (n), as Δ w (n)=- A-1(n) when g (n), e (w (n+1)) obtains minimum value；A (n) is Hessian matrix；
Hessian matrix A (n) is indicated using LM algorithm are as follows:
A (n)=JTJ (11)
Wherein, J is Jacobian matrix；
Gradient vector g (n) is indicated are as follows:
G (n)=JTe(12)
W (n+1) is corrected with following formula:
W (n+1)=w (n)-[JTJ+μI]-1JTe (13)
Wherein, I is unit vector, and μ is constant；
Similarly, the threshold value b (n+1) in (n+1)th iterative process is corrected with following formula:
B (n+1)=b (n)-[JTJ+μI]-1JTe (14)
Wherein, b (n) is the threshold value during nth iteration.
The present invention is by filtering out evaluation index relevant to platform area group of assets comprehensive performance, using based on LM algorithm optimization BP neural network study performance evaluating index and platform area group of assets performance between non-linear relation, and then obtain BP nerve net Network evaluation model passes through the comprehensive performance of BP neural network model evaluation platform area group of assets.
Compared with prior art, the invention has the following advantages:
1) present invention is fitted platform area achievement using the powerful self-learning ability of network, Generalization Ability and Nonlinear Processing ability Complicated non-linear relation between effect and evaluation index, without establishing complicated mathematical model；
2) the platform area group of assets Performance Evaluation Methods proposed by the present invention based on BP neural network have fast convergence, high-precision The advantages that spending.
3) the platform area group of assets Performance Evaluation Methods proposed by the present invention based on BP neural network, which can be evaluated, does not survey Platform area performance score, provide technical support for the performance management of platform area.
Detailed description of the invention
Platform area group of assets Performance Evaluation Methods flow chart in Fig. 1-embodiment of the present invention based on BP neural network；
BP neural network model structure schematic diagram in Fig. 2-embodiment of the present invention.
Specific embodiment
A kind of platform area group of assets Performance Evaluation Methods based on BP neural network, which comprises
1. a pair platform area performance appraisal indices are standardized.Place is standardized to platform area performance evaluating index Before reason, first determining table area Performance Index, as shown in table 1:
Table 1
Platform area performance evaluating index is the input as neural network algorithm, that is, independent variable.Each index value has Different units and magnitude only distinguishes the size of value data for neural network algorithm, not can reflect data Unit.In order to preferably apply above-mentioned algorithm, needs to eliminate the influence of each index Jian Bu commensurate and magnitude logarithm, prevent out The phenomenon that existing " big number eats decimal ".And the standardization of data is exactly to be allowed to fall into one small specific by data bi-directional scaling Section removes the unit limitation of data, is translated into nondimensional pure values.
If platform area number is N, the three-level evaluation index in each area is M, the performance evaluating index group of the area NGe Tai sample Cheng Taiqu performance feature vector, X, has:
Wherein, xijThe i-th row, jth column element for platform area performance feature vector, X, i=1,2 ..., N, j=1,2 ..., M；
Platform area performance evaluating index is standardized, is had:
Wherein, ZijFor xijAmount after standardization,For xijAverage value, sijFor xijVariance.
2. constructing BP neural network evaluation model, and performance value is normalized.Specifically have:
Step 2-1: building BP neural network evaluation model:
BP neural network model includes input layer, hidden layer and output layer, and the transmission function f (a) between each layer is used Logsig function, has:
Wherein, the independent variable of a transmission function f (a) between each layer, 0 < f (a) < 1；
Step 2-2: platform area group of assets items performance evaluating index is normalized:
If platform area performance is d, platform area group of assets performance d is normalized using formula (6), is had:
Wherein, di' for the value after the normalization of i-th area's comprehensive performance, diFor the performance score in i-th area, dminFor The minimum value of all area's performance, dmaxFor the maximum value in all area's performance, α, β are constant, and 0.9 < α <, 1,0 < β < 0.1。
3. according to the comprehensive performance of BP neural network model evaluation platform area group of assets.
The training process of BP neural network platform area group of assets Evaluation Model for Performance, network parameter are provided that the number of iterations 1000, according to network mean square error is according to the minimal error generated in model verification process, it is set as 0.0001, the network number of plies 1 layer, input neuronal quantity is 13, and output neuron quantity is 1, and hidden neuron quantity is 18, initial weight and threshold value with Machine generates.
Algorithm 3-1: the BP neural network training based on the evaluation platform area group of assets performance that matlab code is realized
Algorithm 3-1 is the training process for carrying out Evaluation Model for Performance to platform area group of assets using BP neural network, is first had to Input training data and desired output；Then network parameter is set, which mainly realizes when creating network, main to wrap The quantity for including setting hidden neuron is 18, the quantity of output layer neuron is 1, the form of transfer function be tansig and Logsig, the optimization algorithm used are LM algorithms, are indicated in matlab with the form of trainlm, and minmax function is mainly used It is standardized in input data, newff is for creating neural network and each parameter being arranged；The number of iterations is finally set With defined network error, network is trained using train function, while saving network mode.
Then, the training process of BP neural network platform area group of assets Evaluation Model for Performance: firstly the need of the new test of input Then sample is standardized test sample and carries out ability prediction using trained network.Algorithm 3-2: it is based on The BP neural network model measurement for the evaluation platform area group of assets performance that matlab code is realized
Algorithm 3-2 is the test process for carrying out Evaluation Model for Performance to platform area group of assets using BP neural network, sharp first With load function input test sample test.txt；Then the data newly inputted are evaluated using sim function, sim function Parameter include network mode net and standardization after test sample.
The results show that based on BP neural network evaluation model in the effect that training effectiveness, precision of prediction and sample are verified It meets and performance appraisal is carried out to platform area group of assets, be based particularly on the BP neural network evaluation number prediction rate of LM optimization algorithm Reach 99.5%, illustrates that the platform area group of assets Evaluation Model for Performance based on BP neural network is feasible to the practical problem is solved 's.

Claims (2)

1. a kind of platform area group of assets Performance Evaluation Methods based on BP neural network, it is characterised in that this method includes following step It is rapid:
Step 1: performance appraisal indices in platform area are standardized；
Step 2: building BP neural network evaluation model, and performance value is normalized；
Step 3: according to the comprehensive performance of BP neural network model evaluation platform area group of assets；
In the step 1, if platform area number is N, the three-level evaluation index in each area is M, and the performance of the area NGe Tai sample is commented Valence index forms platform area performance feature vector, X, has:
Wherein, xijThe i-th row, jth column element for platform area performance feature vector, X, i=1,2 ..., N, j=1,2 ..., M；
Platform area performance evaluating index is standardized, is had:
Wherein, ZijFor xijAmount after standardization,For xijAverage value, sijFor xijVariance；
The step 2 the following steps are included:
Step 2-1: building BP neural network evaluation model:
BP neural network model includes input layer, hidden layer and output layer, and the transmission function f (a) between each layer uses logsig Function has:
Wherein, the independent variable of a transmission function f (a) between each layer, 0 < f (a) < 1；
Step 2-2: platform area group of assets items performance evaluating index is normalized:
If platform area performance is d, platform area group of assets performance d is normalized using formula (6), is had:
Wherein, d 'iValue after being normalized for i-th area's comprehensive performance, diFor the performance score in i-th area, dminIt is all The minimum value of platform area performance, dmaxFor the maximum value in all area's performance, α, β are constant, and 0.9 < α <, 1,0 < β < 0.1；
The step 3 the following steps are included:
Step 3-1: using BP neural network model to ZijWith d 'iCarry out learning training；
Step 3-2: platform area performance evaluating index is brought into BP neural network model, calculates platform area comprehensive performance d.
2. the platform area group of assets Performance Evaluation Methods according to claim 1 based on BP neural network, it is characterised in that: institute It states in step 3-1, for any area, since the performance evaluating index in each area is M, so input layer contains M BP Neuron, if input layer input vector is Zr=(Z1,Z2,…,Zm,…,ZM)T, the output vector of hidden layer is Yr=(y1, y2,…,yp,…,yP)T, the output vector of output layer is or=(o1,o2,…,ol,…,oL)T, desired output vector is dr= (d1,d2,…,dl,…,dL)T, wherein T indicates transposition, ZmFor m-th of BP neuron of input layer, ypIt is p-th of hidden layer BP neuron, o1For the 1st neuron of output layer, d1For the 1st desired throughput, P is the BP neuron number of hidden layer, L For the BP neuron number of output layer, m=1,2 ..., M, p=1,2 ..., P, l=1,2 ..., L；Power of the input layer to hidden layer Value and threshold value are respectively wmpAnd bmp, the weight and threshold value of hidden layer to output layer are respectively wplAnd bpl；
Using BP neural network model to ZijWith d 'iThe forward-propagating process for carrying out learning training is as follows:
Output error e is indicated are as follows:
Using BP neural network model to ZijWith d 'iThe back-propagation process for carrying out learning training is as follows:
In (n+1)th iterative process, e is unfolded by Taylor's formula, obtains formula (10):
E (w (n+1))=e (w (n))+gT(n)Δw(n)+0.5ΔwT(n)A(n)Δw(n) (10)
Wherein, w (n) is the weight during nth iteration, and e (w (n)) is output error during nth iteration；w(n+1) For the weight in (n+1)th iterative process, e (w (n+1)) is the output error in (n+1)th iterative process；G (n) be gradient to Amount, T indicate transposition；Δ w (n) is the variable quantity of (n+1)th with weight during nth iteration, i.e. Δ w (n)=w (n+1)-w (n), as Δ w (n)=- A-1(n) when g (n), e (w (n+1)) obtains minimum value；A (n) is Hessian matrix；
Hessian matrix A (n) is indicated using LM algorithm are as follows:
A (n)=JTJ (11)
Wherein, J is Jacobian matrix；
Gradient vector g (n) is indicated are as follows:
G (n)=JTe (12)
W (n+1) is corrected with following formula:
W (n+1)=w (n)-[JTJ+μI]-1JTe (13)
Wherein, I is unit vector, and μ is constant；
Similarly, the threshold value b (n+1) in (n+1)th iterative process is corrected with following formula:
B (n+1)=b (n)-[JTJ+μI]-1JTe (14)
Wherein, b (n) is the threshold value during nth iteration.
CN201811350487.0A 2018-11-14 2018-11-14 A kind of platform area group of assets Performance Evaluation Methods based on BP neural network Pending CN109460925A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811350487.0A CN109460925A (en) 2018-11-14 2018-11-14 A kind of platform area group of assets Performance Evaluation Methods based on BP neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811350487.0A CN109460925A (en) 2018-11-14 2018-11-14 A kind of platform area group of assets Performance Evaluation Methods based on BP neural network

Publications (1)

Publication Number Publication Date
CN109460925A true CN109460925A (en) 2019-03-12

Family

ID=65610303

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811350487.0A Pending CN109460925A (en) 2018-11-14 2018-11-14 A kind of platform area group of assets Performance Evaluation Methods based on BP neural network

Country Status (1)

Country Link
CN (1) CN109460925A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110675017A (en) * 2019-08-13 2020-01-10 平安科技（深圳）有限公司 Performance evaluation method and device based on artificial intelligence
CN111783987A (en) * 2020-07-14 2020-10-16 中国水利水电科学研究院 Farmland reference crop evapotranspiration prediction method based on improved BP neural network

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105590175A (en) * 2016-02-15 2016-05-18 云南电网有限责任公司 Skilled talent evaluation method based on factor analysis and BP neural networks
CN106127387A (en) * 2016-06-24 2016-11-16 中国电力科学研究院 A kind of platform district based on BP neutral net line loss per unit appraisal procedure
US20180246853A1 (en) * 2017-02-28 2018-08-30 Microsoft Technology Licensing, Llc Hardware node with matrix-vector multiply tiles for neural network processing

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105590175A (en) * 2016-02-15 2016-05-18 云南电网有限责任公司 Skilled talent evaluation method based on factor analysis and BP neural networks
CN106127387A (en) * 2016-06-24 2016-11-16 中国电力科学研究院 A kind of platform district based on BP neutral net line loss per unit appraisal procedure
US20180246853A1 (en) * 2017-02-28 2018-08-30 Microsoft Technology Licensing, Llc Hardware node with matrix-vector multiply tiles for neural network processing

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110675017A (en) * 2019-08-13 2020-01-10 平安科技（深圳）有限公司 Performance evaluation method and device based on artificial intelligence
CN111783987A (en) * 2020-07-14 2020-10-16 中国水利水电科学研究院 Farmland reference crop evapotranspiration prediction method based on improved BP neural network

Similar Documents

Publication Publication Date Title
Cai et al. Exploration on the financing risks of enterprise supply chain using Back Propagation neural network
CN106127387A (en) A kind of platform district based on BP neutral net line loss per unit appraisal procedure
CN105786681A (en) Server performance evaluating and server updating method for data center
CN105279692A (en) Financial information technology system performance prediction method and apparatus
CN107437111A (en) Data processing method, medium, device and computing device based on neutral net
CN109460925A (en) A kind of platform area group of assets Performance Evaluation Methods based on BP neural network
CN106203616A (en) Neural network model training devices and method
Popkova et al. Energy efficiency and pollution control through ICTs for sustainable development
CN112700326A (en) Credit default prediction method for optimizing BP neural network based on Grey wolf algorithm
CN113378467B (en) Method and device for fiber channel modeling, electronic equipment and storage medium
Pan et al. Near‐optimal control of a stochastic vegetation‐water system with reaction diffusion
CN110163419A (en) A kind of method of middle and small river river basin flood forecast
CN107391442A (en) A kind of augmentation linear model and its application process
CN103763123A (en) Method and device for evaluating health condition of network
Zhu et al. Selection of criteria for multi-criteria decision making of reservoir flood control operation
Liu et al. Analysis of efficiency of human resource management evaluation model based on SOM neural network
Pang et al. QML-ainet: An immune network approach to learning qualitative differential equation models
CN111160662A (en) Risk prediction method, electronic equipment and storage medium
Li et al. Construction of College Students’ Employment Quality Evaluation Model System under the Background of Digitalization
Seth Use of artificial neural networks and genetic algorithms in urban water management: a brief overview
Mirzaev et al. A numerical framework for computing steady states of size-structured population models and their stability
Qi et al. A Bio-Inspired Algorithm for Maximum Matching in Bipartite Graphs.
Xu et al. BP neural network-based product quality risk prediction
Motameni et al. Lookback option pricing under the double Heston model using a deep learning algorithm
Xuefang et al. Study on the risk assessment of real estate project based on BP neural network

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20190312