Download - Research Article Measuring Technological Progress of Smart ...downloads.hindawi.com/journals/mpe/2014/861820.pdfResearch Article Measuring Technological Progress of Smart Grid Based

Research ArticleMeasuring Technological Progress of Smart Grid Based onProduction Function Approach

Dong Han and Zheng Yan

Key Laboratory of Control of Power Transmission and Conversion, Department of Electrical Engineering, Shanghai Jiao TongUniversity, Ministry of Education, Dianxinqunlou 1-127, Dongchuan Road No. 800, Minhang District, Shanghai 200240, China

Correspondence should be addressed to Dong Han; [email protected]

Received 24 January 2014; Revised 21 July 2014; Accepted 20 August 2014; Published 25 September 2014

Academic Editor: Minrui Fei

Copyright © 2014 D. Han and Z. Yan. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Production function theory combined with data envelopment analysis (DEA) and ridge regression analysis (RRA) is applied toevaluate the technological progress of the smart grid.The feasible conditions of production function models are determined by theDEA algorithm. RRA is applied to estimate the relevant parameters of the evaluation models under study. One of the significantsteps in the design of the assessment algorithm is the structure of production function models. Therefore, the Cobb-Douglas,constant elasticity of substitution, and translog production functions are employed to evaluate the technological progress of thesmart grid, respectively. The results of analysis and calculation mainly include the DEA relative efficiency, slacks in inputs andoutputs of inefficient units, estimated parameters, and quantitative indices of technological progress.

1. Introduction

The smart grid is a new modern power grid, and it ownsadvanced metering technologies, information communica-tion technologies, analysis and decision technologies, auto-matic control technologies, and highly integrated physicalinfrastructures [1]. Different from traditional power grid, theintelligence is the most significant attribute, and it is also thecore value of the smart grid, to improve the socioeconomicbenefits for the public. Generally, the intelligent technologiesof the smart grid mainly include advanced technologiesand equipment in the generation, transmission, substation,distribution, and dispatching fields, and they will enhancethe self-healing ability, the integration of information andcommunication, the highly efficiency of management, andthe interaction with consumers to play a part in optimizingthe operation of the system [2].

The evaluation of technological progress not only will beable to reflect the technological level of a smart grid, but alsocan measure the economic benefits brought from the appliedadvanced technologies. However, the smart grid as a com-prehensive engineering is with a long construction period,intensive investments, and highly technical difficulties. It isvery hard to quantitatively identify the development level of

the smart grid. Hence, how to evaluate the construction effectof a smart grid and the intelligent technology availabilityhas become one of the challenges for the current assessmentresearch of the smart grid. It is necessary to present anevaluationmethodology tomeasure technological progress ofa smart grid.

Although the construction state of smart grid is stillin the initial stage, the evaluation research and practice ofsmart grid have been reported preliminarily. The US ElectricPower Research Institute (EPRI) designed the assessmentsystem for smart grid programs over the planning andconstruction periods, for the purpose of the identificationof the technological levels and the metrics of smart grids.Furthermore, this work would be helpful to perform the cost-benefit analysis of smart grids in US [3, 4]. Different from theEPRI, the USDepartment of Energy only outlined the overalldevelopment ideas and somemajormetrics for the smart gridin the report, not describing the specific interpretations indetail [5]. The European Network of Transmission SystemOperators (ENTSO) also conducted an evaluation indexsystem of investment grant project for the European smartgrid, but not analyzing the technological benefits [6].

As known to all, the smart grid has been a hot topicin electrical engineering sector. An effective and scientific

Hindawi Publishing CorporationMathematical Problems in EngineeringVolume 2014, Article ID 861820, 9 pageshttp://dx.doi.org/10.1155/2014/861820

2 Mathematical Problems in Engineering

evaluation method is beneficial to the identify, the prob-lems of smart grid construction, and advanced technologyapplication. Hence, in this paper we attempt to present amethodology to evaluate the technical level of the smartgrid based on the product function.The production functionspecifies the maximum output that can be produced witha given quantity of inputs. It is defined for a given state ofengineering and technical knowledge. From the economicpoint of view, technology innovation occurs when newengineering knowledge improves production techniques forexisting products. Such technological change is equivalentto a shift in the production function. Consequently, theproduction function models are developed to assess thetechnological progress and socioeconomic benefits of thesmart grid.Moreover, we consider the application of the DEAmethodology to determine the efficient production state ofthe smart grid, because efficient production is the necessarycondition of application to production function theory [7–9]. Through the analysis of DEA, the production functionmodels are built based on the output, input, and technologyitems. The advantage of the proposed methodology is toquantitatively evaluate the impact of the smart grid technolo-gies on economic benefits, which will show the intellectual-ization effects of a power grid. In addition, ridge regression isemployed to estimate the parameters of production functionmodels. Finally, case studies demonstrate the effectiveness ofthe proposed approaches [10].

This paper is organized as follows. Section 2 constructssome mathematical models with DEA and production func-tion theory. In Section 3, the application of the proposedmethodology is presented. Some discussion about the prop-erty of production function is done in Section 4. Finally,Section 5 summarizes the main conclusions and contribu-tions of this paper.

2. Methodology

2.1. Data Envelopment Analysis. The DEA is an efficiencymodeling approach that can be widely used to measure therelative efficiency of different decision-making units (DMUs).The DEA can not only analyze the simple input-output ratio,but also handle multiple input-output variables. The purposeof applying the DEA is to provide a judging standard thatshows that the production state for the smart grid is efficient,and then the production function can be used to analyze thetechnological change. Otherwise, supposing the productionbased on the given inputs and outputs is inefficient, the slackanalysis of DEA will offer the improved measure enablingthe efficient perfect input-to-output state. Mathematically,the DEA algorithm is in essence a linear programmingprocedure. The formulation for the DEA methodology canbe described as follows:

max 𝜇𝑇𝐵0

s.t. 𝑤𝑇𝐴𝑗− 𝜇𝑇𝐵𝑗≥ 0

𝑤𝑇𝐴0= 1

𝑤 ≥ 𝑒𝑠, 𝜇 ≥ 𝜀𝑒

𝑚

𝑗 = 1, 2, . . . , 𝑛,

(1)

where 𝜇 and 𝑤 are, respectively, the weight coefficients ofinput and output variables,𝐴

𝑗is the amount of input utilized

by the DMU 𝑗, 𝐵𝑗is the amount of output produced by DMU

𝑗, the notation 0 is the designed unit for an optimization run,𝜀 is a small positive number, 𝑒

𝑠is the 𝑠 dimension unit vector,

and 𝑒𝑚is the𝑚 dimension unit vector.

The above model is Charnes-Cooper-Rhodes (CCR)model that is suitable for DEA-based study of electric utilities[11]. The CCR model is first presented and named by theoperation researchers of A. Charnes, W. W. Cooper, andE. Rhodes. It is remarkable that the CCR model is linearprogramming problem in mathematics. Instead of solvingthe CCR model as stated above, an equivalent model ispresented since it requires lesser computations and is easierto implement.The dual program formof primaryCCRmodelis represented as follows:

max 𝜃 − 𝜀 (e𝑇𝑠𝑆 + e𝑇𝑚𝑀)

s.t.𝑛

∑

𝑗=1

𝜆𝑗𝐴𝑗+ 𝑆 = 𝜃𝐴

0

𝑛

∑

𝑗=1

𝜆𝑗𝐵𝑗+ 𝑆 = 𝜃𝐴

0

𝜆𝑗≥ 0, 𝑗 = 1, . . . , 𝑛

𝑆 ≥ 0, 𝑀 ≥ 0,

(2)

where 𝜃 is the scalar quantity that is technical efficiency score,𝜆𝑗is the decision variable of the DMU 𝑗, and 𝑆 and 𝑀 are,

respectively, the slack and surplus variables.Once the optimal solution 𝜃

∗= 1, 𝑆∗ = 0, 𝑀

∗= 0,

it illustrates that the DMU is called DEA efficient [12, 13].The slack vectors including the input excess and the outputshortfalls are defined as

Δ𝐴 = 𝐴 − (𝜃∗𝐴 − 𝑆)

Δ𝐵 = 𝐵 +𝑀,

(3)

where Δ𝐴 is the gap vector in inputs, Δ𝐵 is the gap vector inoutputs, 𝐴 is the input vector, and 𝐵 is the output vector.

2.2. Production Function Theory. The production functionfocuses on the relationship between the amount of inputrequired and the amount of output that can be obtained.Suppose 𝑌 is the output vector and 𝑥

1, 𝑥2, . . . , 𝑥

𝑛is the

combination of input variables; the production function canbe generally described as

𝑌 = 𝐹 (𝑥1, 𝑥2, . . . , 𝑥

𝑛) . (4)

It is noticeable that technological progress is an implicitvariable and it is difficult to be calculated by the universalmethods. Based on the idea of “residual value,” the technologyas the independent variable can be separated from theproduction function. Hence, technological progress regardedas a residual value is calculated indirectly by this way. Sometypical production functions are introduced as follows.

2.2.1. Cobb-Douglas (C-D) Production Function. The C-Dspecification is described as a function with the input

Mathematical Problems in Engineering 3

and maximum amount of output that can be producedusing a combination of applied production technology [14].The inputs consist of the capital investment and the laborresource. The mathematical expression of a C-D productionfunction is presented in [15] as follows:

𝑌 = 𝐴𝐾𝛼𝐿𝛽, 0 < 𝛼, 𝛽 < 1, (5)

where 𝐴 is the technological progress variable, 𝐾 is thecapital investment variable, 𝐿 is the labor resource variable,𝑌 is the output variable, and 𝛼 and 𝛽 are, respectively, theoutput elasticities of capital and labor. It is remarkable thatsome assumptions play a key role in the derivation of C-Dproduction function: constant returns to scale and perfectcompetition. Under the law of constant returns to scale, thesum of 𝛼 and 𝛽 is equal to one. Moreover, it is also assumedthat the technological progress for the production is neutral.

2.2.2. Constant Elasticity of Substitution (CES) ProductionFunction. The classical CES production function derived byArrow, Chenery, Minhas, and Solow in 1961 is one of themost widely used production functions.The CES productionfunction is developed based on the assumption that therelationship between 𝑌/𝐿 (output per unit of labor) and𝑊 (the wage rates) is independent of the stock of capital.However, the CES production function is also subject to thelimitation that the value of the elasticity of substitution isconstant although not necessarily equal to one. The explicitformula of a CES production function is described in [16]

𝑌 = 𝐴[𝛿𝐾−𝑝

+ (1 − 𝛿) 𝐿−𝑝]−𝑟/𝑝

0 < 𝛿 < 1, −1 < 𝑝 < ∞,

(6)

where 𝛿 is the proportional distribution parameter, 𝑟 is thescale parameter, 𝑟 > 1, 𝑟 = 1, or 𝑟 < 1, respectively,corresponds to increasing returns to scale, constant returnsto scale, or decreasing returns to scale, and 𝑝 is the substi-tution parameter. In particular, while 𝑝 tends to zero, theCES production function will be transformed into the C-D production function, so C-D production function is thespecial form of CES production function.

2.2.3. Translog Production Function. The translog productionfunction imposes nomore restrictions on returns to scale andthe elasticity of substitution than the production functionsabove. The translog production function is recommendedin [17], and the mathematical representation is defined asfollows:

ln𝑌 = 𝛼0+ 𝛼𝐾ln𝐾 + 𝛼

𝐿ln 𝐿 + 𝛼

𝑇ln𝑇

+1

2𝛽𝐾𝐾

ln2𝐾 + 𝛽𝐾𝐿

ln𝐾 ln 𝐿

+ 𝛽𝐾𝑇 (ln𝐾)𝑇 +

1

2𝛽𝐿𝐿ln2𝐿

+ 𝛽𝐿𝑇 (ln 𝐿) 𝑇 +

1

2𝛽𝑇𝑇𝑇2,

(7)

where 𝛼0, 𝛼𝐾, 𝛼𝐿, 𝛼𝑇, 𝛽𝐾𝐾

, 𝛽𝐾𝐿, 𝛽𝐾𝑇

, 𝛽𝐿𝐿, 𝛽𝐿𝑇, and 𝛽

𝑇𝑇are the

undetermined parameters and 𝑇 is the time variable.

A major advantage of the translog production function isthat the elasticity of substitution for each input componentis variable. Besides, the translog production function enablesa richer specification of the relationships for the inputscompared to other production functions in the previousdescription. Nevertheless, the translog production functionowns more parameters than the C-D and CES productionfunctions, which means that the complexity of parameterestimation for translog production function will make asignificant challenge.

2.3. Parameter Estimation. Solving the parameter estimationproblem is one of the most significant steps in the evaluationprocedure using the production functions [18]. Consider-ing the characteristics of the estimated parameters in theproposed production functions, such as the collinearity andcorrelation properties, the ridge regression analysis (RRA)is adopted to perform the parameter estimation in thispaper. The RRA is a dedicated to the analysis of the dataof linear biased estimation regression method, and it is inessence an improved least-square estimation method [19].Therefore, it is suitable to employ the RRA to minimize thecorrelation effects of the variables.The fundamental principleof parameter estimation by the RRA is shown in brief [19].

Give the linear model

𝑐 = 𝐺𝑏 + 𝜉, (8)

where 𝐺 is the variable matrix, 𝑐 is the observed vector, 𝑏 isthe estimated parameter, and 𝜉 is the error term. The ridgeestimator of 𝑏 is

�̂� = (𝐺𝑇𝐺 + 𝑘𝐼)

−1

𝐺𝑇𝑐, (9)

where 𝐺𝑇 is the transposed matrix of 𝐺, 𝐼 is the identity

matrix, and 𝑘 is the scalar parameter.

2.4. Quantity Property of Technological Progress. Using theproduction functions, some quantitative indices representingtechnological progress in the production process need to becalculated. The evaluation indices are described in detail asfollows.

2.4.1. Rate of Technical Progress. Therate of technical progressdenotes the effect of saving inputs per unit output in theassessment period. The derivation of the index is generallyintroduced as follows. Consider the following form of thegeneral production function:

𝑌 = 𝐹 (𝑥1, 𝑥2, . . . , 𝑥

𝑛, 𝑡) . (10)

The differential form of (10) is obtained as follows:

𝑑𝑌

𝑌=

𝜕𝐹 (𝑥1, 𝑥2, . . . , 𝑥

𝑛, 𝑡)

𝜕𝑡

𝑑𝑡

𝑌

+

𝑛

∑

𝑖=1

[𝜕𝐹 (𝑥1, 𝑥2, . . . , 𝑥

𝑛, 𝑡) /𝜕𝑥

𝑖

𝐹/𝑥𝑖

]𝑑𝑥𝑖

𝑥𝑖

= 𝑎 (𝑡) +

𝑛

∑

𝑖=1

𝛼𝑖

𝑑𝑥𝑖

𝑥𝑖

,

(11)


where 𝑑𝑌/𝑌 = 𝑦 that is the growth rate of outputs and 𝑑𝑥𝑖/𝑥𝑖

is the input growth rate of the element 𝑖. Then the definitionof rate of technical progress is

𝑎 = 𝑦 −

𝑛

∑

𝑖=1

𝛼𝑖

𝑑𝑥𝑖

𝑥𝑖

. (12)

2.4.2. Technical Contribution to Output Growth. The techni-cal contribution to output growth 𝑆

𝑎is the proportion of the

rate of technical progress in the output growth speed. Thedefinition of this index is represented as follows:

𝑆𝑎=𝑎

𝑦, (13)

where 𝑦 is the actual output growth speed.

2.4.3. TechnicalMerit. Thetechnicalmerit index indicates thetechnological level of the smart grid and it can be measuredby the following form:

𝐴 (𝑡) = 𝐴 (𝑡 − 1) [1 + 𝑎 (𝑡 − 1)] , (14)

where 𝑎(𝑡 − 1) is the growth rate of technical progress at thetime point 𝑡 − 1.

3. Application

In this section, the application of production function theorycombined with DEA and RRA will be implemented. Figure 1shows the overall evaluation process, in which the techno-logical level of smart grid technologies and the technologicalprogress of the smart grid can be displayed by means of theselected evaluation indices.

The smart grid associated with a group of various tech-nologies, attributes, and objectives covers comprehensiveconstruction, where major breakthroughs in key technologyand equipment should be achieved. It is a challenge toevaluate the technical level of a smart grid considering allconcerns. Thus, it is suitable to select a specific attribute orgoal of smart grid to study in detail. One of the representativeobjectives is integrating more clean energy, including solarandwind energy, into electric power grids, which is also takenas a classic example to implement the evaluation of tech-nology level and technological progress for the integrationof clean energy in this paper. The integration of large-scaleclean energy is an important part of smart grid technologies.Generally, the clean energy turbine technology, integrationtechnology into power grids, bulk storage devices, and powerforecast technology have a significant impact on the cleanenergy development. Specifically, a highly efficient cleanenergy turbine can reduce cost and improve reliability. Theoptimal operation and sustainable construction are regardedas an effective suggestion for the improvement of clean energyintegrated into power systems. The flexible bulk storagedevices and the power forecast technologymay overcome thevolatility of renewable energy, so as to promote the utility ofthe clean energy widely. Therefore, to adapt the development

of clean energy in the smart grid, the input-output relation-ship of such objective and the development level of intelligenttechnologies will be analyzed deeply.

The data about the clean energy development plan ina regional power grid has been obtained in Table 1 whichincludes the forecast values of output and input variables inthe next decade. The inputs contain labor and three kindsof investments which are the clean energy investments incapacities (CEIC), the bulk energy storage devices (BESD),and the construction investments of power grids (CIPG).Theoutputs include the reduced paying carbon taxes (RPCT),the benefits from the reduced fossil energy (BRFE), and theelectricity sales of the clean energy (ESCE).

According to the data about the clean energy devel-opment plan in Table 1, the feasibility of the productionfunctions should be analyzed by the DEA technique. TheDEA optimizationmodel is solved byMATLAB optimizationtoolbox. The optimal solution of relative technical efficiencyis gained and the performance of intelligent technologies canalso be understood. Figure 2 shows the results of relativetechnical efficiency at the time points. It turns out that therelations between the inputs and outputs are DEA efficientin years 1, 2, 3, 5, 8, 9, and 10, respectively. Furthermore, italso illustrates that the technical efficiencies are available andthe returns to scale are constant in these years. However, theratio efficiency between the inputs and outputs is inefficientin years 4, 6, and 7. In order to identify the reason why therelative technical efficiency is not available, it is necessaryto analyze the gaps of the outputs and inputs. The analysiswill provide the suggestions of how to adjust the originaloutputs and inputs, so that the application of the productionfunctions can be achieved. Figure 3 shows the state of slackvariables of the inputs for power grids in the evaluation cycle.In Figure 3, it shows that the slack variables of investment 1,investment 2, and investment 3 are unequal to zero, whichmeans that the investments are obviously in the idle states andthe values of slack variables are equal to the idle quantity ofthe investments of power grids.

The parameter estimation is also a significant step inoverall procedure of the technological progress evaluation.Due to the advantage of RRA that is capable of coping withthe problem of collinearity between each variable in theproduction functions, the results of the estimated parameterswill be more accurate using the RRA technique. Tables 2and 3 show the estimated parameter values using the C-D,CES, and translog production functions, respectively, withthe scalar parameter of ridge regression 𝑘 = 0.11, 0.13, and0.2. All the numerical results in Tables 2 and 3 are calculatedby the SPSS software based on the observed measures of theinput and output historical data [20]. The results in Table 2indicate that the economic scale is the constant returns toscale, owing to the equation of 𝛼 + 𝛽 = 1 and 𝑟 ≈ 1.The results in Table 3 demonstrate that all the coefficients ofthe translog production function are positive, which meansthat the technical progress is neutral and the trend will beaccelerating in the evaluation cycle. In addition, the resultsfrom SPSS show 𝑅

2= 0.987, 0.951, and 0.865 for the C-D,

CES, and translog production models, which indicates that


Start

Variables initialization

Suppose t = 1

Statistics of outputs,investment, and labor

DEA is availableor not?

Calculation of average annual growingrate of economic variables

Calculation of technologicaldevelopment speed and economic

benefits

Calculation of technological progressindices

Terminate

Adjustment?

t = t + 1

No

Yes

No

Yes

Yes

No

Parameters estimation

t = tmax?

Figure 1: Block diagram of technological progress evaluation.

the complex determined coefficients𝑅2 are highly significant.Consequently, these estimated parameters are reasonable andnearly conform to the actual condition of the smart grid aswell.

Through theDEA examination analysis ofmultiple input-output variables and the parameter estimation of the produc-tion functions, the assessment of technological progress of thesmart grid can be performed. The rate of technical progress,technical contribution to output growth, and technical merit,representing the indices of the technological progress level ofthe smart grid, can be calculated by the proposed production

functions in the evaluation cycle. Figures 4 and 5 show theindex results about the rate of technical progress and tech-nical contribution to output growth, respectively. The resultsin Figure 4 indicate that the values of the rate of technicalprogress obtained from the translog production function aresignificantly smaller than other production functions. Exceptthe results in the second year and the last year, the index val-ues calculated by the CES and C-D production functions areapproximately the same. The results in Figure 5 demonstratethat the numerical values of the technical contribution tooutput growth calculated by the translog production function


Table 1: The input data of clean energy development plan.

Time (year) Investment1: CEIC

Investment2: BESD

Investment3: CIPG Labor force Income

1: RPCTIncome2: BRFE

Income3: ESCE

1 23.8 6.50 2.50 1.49 15.6 4.00 23.402 25.1 7.20 3.00 1.50 16.1 4.80 24.803 27.6 7.59 3.10 1.51 16.8 5.42 26.134 30.4 8.23 3.40 1.54 17.5 6.09 27.535 32.5 8.56 3.60 1.55 18.8 6.77 29.676 35.4 9.22 3.91 1.57 19.1 7.54 30.707 37.7 9.71 4.10 1.59 19.3 7.93 31.248 38.5 9.51 4.25 1.61 20.1 8.27 32.549 39.3 10.11 4.44 1.65 21.0 8.82 34.1210 42.1 11.00 4.82 1.64 22.5 9.80 36.80The units of measurements of investment and income 1, 2, and 3 are hundred million dollars. The unit of measurement of labor force is ten thousand people.

1 2 3 4 5 6 7 8 9 100.96

0.965

0.97

0.975

0.98

0.985

0.99

0.995

1

1.005

1.01

Relat

ive t

echn

ical

effici

ency

Time (year)

Figure 2: Input-output relative technical efficiency for power gridsin evaluation cycle.

Table 2: The results of parameter estimation of C-D and CESproduction functions.

Parameter Result Standard error𝛼 0.7321 0.8103𝛽 0.2679 0.2017𝑟 0.9923 1.1205𝑝 0.5236 0.5138𝛿 0.7401 0.6942

are distinctly less than the results obtained from others.Moreover, the difference between the translog productionfunction and other production functions in the initial stage isslightly bigger than its later stage.With respect to the propertyof the rate of technical progress, the values calculated by theCES and C-D production functions are also approximatelyuniform. Synthesizing the data analysis for Figures 4 and5, we can generally summarize that the calculation resultsobtained from the translog production function tend to

1 2 3 4 5 6 7 8 9 100

0.5

1

1.5

2

2.5

3

Time (year)

Slac

k va

riabl

es

Investment 1Investment 2

Labor forceInvestment 3

Figure 3: Schematic drawing of slack variables of inputs for powergrids.

be rather conservative, while the calculation results of theC-D and CES production functions display the optimisticproperty. Another point of view from the data analysis isthat the calculation results obtained from theCES productionfunction are close to the results of the C-D productionfunction. The main reason is that on one hand they have thesimilar function expression and on the other hand they ownthe same application condition that the environment is theconstant returns to scale.

The index values of the technical merit are shown inFigure 6, and the results indicate that the technical merit ofthe smart grid is improved annually in the evaluation periods.The growth pattern of the technical merit is similar to anexponential function form. Due to the significant differenceof the technical merit between the translog production func-tion and other production functions, the numerical results


Table 3: The results of parameter estimation of translog production functions.

Parameter Result Standard error Parameter Result Standard error𝛼0

0.0802 0.0928 𝛽𝐾𝐿

0.0100 0.0236𝛼𝐾

0.2539 0.3641 𝛽𝐾𝑇

0.0013 0.0093𝛼𝐿

0.0316 0.0472 𝛼0

0.0802 0.0928𝛼𝑇

0.0195 0.0307 𝛼𝐾

0.2539 0.3641𝛽𝐾𝐾

0.0402 0.0583 𝛼𝐿

0.0316 0.0472

2 3 4 5 6 7 8 9 100

2

4

6

8

10

12

14

16

18

20

22

Time (year)

Rate

of t

echn

ical

pro

gres

s (%

)

Translog production functionCES production functionC-D production function

Figure 4: The rate of technical progress calculated by productionfunctions.

of the translog production function are also less than theresults of others. The technical merit results reveal that thetechnological level of smart grid technologies is enhancedyearly. It alsomeans that intelligent technologies are generallyused widely in the smart grid.

The indices of the rate of technical progress, technicalcontribution to output growth, and technical merit representthe intelligent properties of the smart grid. Especially, the dataresults in Figure 5 show that the technological proportionis from about 20% to 30% for the C-D production func-tion, which illustrates that the revenues from the intelligenttechnology are less than the investment and labor. It isnecessary for managers to take effective measures to improvethe intelligent level of the smart grid, furthermore promotingthe extensive application of intelligent technologies.

As for the proposed assessment models based on the pro-duction function approaches in this paper, a more importantquestion is how to choose from these production functionsto reflect the actual smart grid. In authors’ opinion, it seemsthat the C-D production function could be used to evaluatethe technological progress more properly in most cases. Thereason is that the CES and translog production functions

2 3 4 5 6 7 8 9 100

5

10

15

20

25

30

35

Time (year)

Tech

nica

l con

trib

utio

n to

out

put g

row

th (%

)

C-D production functionCES production functionTranslog production function

Figure 5:The technical contribution to output growth calculated byproduction functions.

1 2 3 4 5 6 7 8 9 101

1.2

1.4

1.6

1.8

2

2.2

2.4

2.6

2.8

Time (year)

Tech

nica

l mer

it

C-D production functionCES production functionTranslog production function

Figure 6: The technical merit calculated by production functions.


2 3 4 5 6 7 8 9 108

10

12

14

16

18

20

22

24

26

28

Time (year)

Tech

nica

l con

trib

utio

n to

out

put g

row

th (%

)

Translog production function ITranslog production function II

Figure 7: Technical contribution to output growth calculated bytranslog production models.

require more complex computation to estimate their param-eters, which may result in the deterioration of calculationprecision. Although a critical application condition of theC-D production function is limited to constant returns toscale, most actual situations in not only power systems butalso other industries are thought to submit to it in general.However, if the decision-maker tends to implement morecomplex and detailed analysis to evaluate the technologicalprogress, the CES and translog production functions will berecommended. Under the same conditions, C-D productionfunction can be appliedmorewidely and conveniently. Undersome special circumstances, it should be noted that differentproduction models have their corresponding and uniqueapplied scopes, which will be discussed in the next section.

4. Discussion

As to the mathematical formulations of the productionfunctions, the structure of the C-D production function issimilar to the CES production function and they both havethe unique inputs such as technology, capital investment, andlabor. Moreover, the C-D and CES production functions canbe widely used because of the simple expressions. However,for the translog production function, though it has morecomplex mathematical representation than other productionfunctions, the law of restricting the fixed inputs and theconstant elasticity of substitution may not be complied.Therefore, the translog production function can be appliedon more academic areas considering more and more non-routine factors. For example, the environmental factor is animportant objective for smart grid development. Supposingthe additional load demand as an environmental factor is

contained in the new translog production function, thespecific formulation is introduced as follows:

ln𝑌 = 𝛽0+ 𝛽𝐾ln𝐾 + 𝛽

𝐿ln 𝐿 + 𝛽

𝐷ln𝐷 + 𝛽

𝑇ln𝑇

+ 𝛼𝐾𝐿

ln𝐾 ln 𝐿 + 𝛼𝐾𝐷

ln𝐾 ln𝐷 + 𝛼𝐿𝐷

ln 𝐿 ln𝐷

+1

2𝛼𝐾𝐾

ln2𝐾 +1

2𝛼𝐿𝐿ln2𝐿 + 1

2𝛼𝐷𝐷

ln2𝐷 +1

2𝛼𝑇𝑇ln2𝑇

+ 𝛼𝐾𝑇 (ln𝐾)𝑇 + 𝛼

𝐿𝑇 (ln 𝐿) 𝑇 + 𝛼𝐷𝑇 (ln𝐷)𝑇,

(15)

where𝐷 is the additional load demand.The additional load demand can be approximately fore-

casted according to the smart grid plans, technological inno-vation, and policy orientations. In [21], the forecast resultsof the additional load demand are given. Based on the dataof the forecast additional load demand, the index values oftechnical contribution to output growth obtained from thedifferent production functions can be shown in Figure 7.Translog production function I includes the additional loaddemand factor, while translog production function II is theoriginal form of the translog production function previouslymentioned. The difference from the results indicates that theinfluence of additional load demand cannot be ignored. Inother words, the additional factors may conduct the differentindex results to represent technological level; consequently, itis necessary to focus on the impact of the various factors onthe technological progress assessment of the smart grid.

For the proposed methodology to evaluate the techno-logical progress of smart grids, it describes an empiricalrelationship between specific output and inputs for a powergrid. In the modeling process, the production functions areused to represent the output production generated frominvestment and labor inputs, as well as technology. Forapplication to production function approaches, we assumethat the input variables include 𝑌, 𝐿, and 𝐾 and the outputvariables are 𝑎 and 𝑆

𝑎. Hence, the technological progress

of smart grids can be measured by this measure that is aparametric method in operations research and economicsfor the estimation of power system production. In addition,DEA technique as a nonparametric method is used to selectthe optimal inputs and output of production functions. DEAis a preprocessing using, underlying the application to pro-duction function approaches.The evaluation of technologicalprogress proposes such mathematical problem that is a time-series estimation of the production state of smart grids basedon multiple inputs/outputs in power system planning andoperation models. In the solving procedure, the followingproperties of the evaluation framework can be obtained. (i)The models approached by production functions are builton the assumption of DEA availability. (ii) Besides the inputvariables 𝑌, 𝐿, and 𝐾 determined by DEA, the parameters 𝛼and𝛽 also have the impact on the results of technology assess-ment. (iii) Data for a portion of the technological progressevaluation can provide the primary basis for explorationof the production function model, while the data used toimplement the assessment can play a significant role in theaccurate estimation.


The evaluation framework can perform the technologicalprogress assessment for smart grids from a macroview.Moreover, we assume that technical contribution to out-put growth represents technological progress in evaluationmodels. Different from the commonly evaluation methods,such as comprehensive assessment approach and cost-benefitanalysis, this paper presents a novel evaluation model basedon the parametric and nonparametric estimation methodsto implement technology-based assessment for smart grids.The proposed methodologies can be used to evaluate theeffects of the adopted intelligent technologies in smart gridconstruction, which is helpful to direct the future powersystem planning and operation.

5. Conclusion

This paper presents the evaluation methodology to measurethe technological progress of the smart grid based on produc-tion function theory.Theproposedmethod ismathematicallyformulated to analyze the relationship betweenmultiple inputand output variables of the smart grid. In the evaluationprocess, the DEA test is regarded as an important step toensure the application condition of the production functionsin economic law. The indices representing the technologicalprogress characteristic of the smart gird are obtained fromthe adopted C-D, CES, and translog production. Moreover,the simulation results in case studies indicate that thetendency of technological levels is generally improved. Thecomparison analysis about the different production functionsis performed in discussion, from which the applicationscope, modeling mechanism and engineering value of theproduction function can be understood. Finally, this study isa first strategic approach for the evaluation of technologicalprogress of the smart grid from the macro view.

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper.

Acknowledgments

The authors would like to thank the State Energy SmartGrid R&D Center (Shanghai) for the technical support. Inaddition, this work was supported by the National NaturalScience Foundation of China (Grant no. 51377103).

References

[1] J. D. McDonald, “The next-generation grid : EEEnergy infras-tructure of the future,” IEEE Power and Energy Magazine, vol. 7,no. 2, pp. 26–94, 2009.

[2] M. Amin, “Toward self-healing energy infrastructure systems,”IEEE Computer Applications in Power, vol. 14, no. 1, pp. 20–28,2001.

[3] W. Wakefield, “Methodological approach for estimating thebenefits and costs of smart grid demonstration projects,”U.S.ERPI, Palo Alto, Calif, 2010.

[4] Smart grid characteristics, values, and metrics, http://energy.gov/.

[5] U. S. Department of Energy, Smart Grid System Report, U. S.DOE, Washington, DC, USA, 2009.

[6] European Commission, “European technology platform smartgrids: vision and strategy for Europe’s electricity networks of thefuture,” European Commission, 2010.

[7] F. Y. Lo, C. F. Chien, and J. T. Lin, “A DEA study to evaluatethe relative efficiency and investigate the district reorganizationof the Taiwan power company,” IEEE Transactions on PowerSystems, vol. 2, no. 16, pp. 170–178, 2001.

[8] L.D.Otero, G.Centeno, C. E.Otero, andK. Reeves, “ADEA-tobitanalysis to understand the role of experience and task factorsin the efficiency of software engineers,” IEEE Transactions onEngineering Management, vol. 59, no. 3, pp. 391–400, 2012.

[9] T. Schoenherr and S. Talluri, “Environmental sustainabilityinitiatives: a comparative analysis of plant efficiencies in Europeand the U.S.,” IEEE Transactions on Engineering Management,vol. 60, no. 2, pp. 353–365, 2013.

[10] B. Jayasekara and U. D. Annakkage, “Derivation of an accuratepolynomial representation of the transient stability boundary,”IEEE Transactions on Power Systems, vol. 21, no. 4, pp. 1856–1863, 2006.

[11] B. Golany, Y. Roll, and D. Rybak, “Measuring efficiency ofpower plantsin Israel by data envelopment analysis,” IEEETransactions on Energy Management, vol. 8, no. 41, pp. 291–301,1994.

[12] A. Pahwa, X. Feng, and D. Lubkema, “Performance evaluationof electric distribution utilities based on data envelopmentanalysis,” IEEE Transactions on Power Systems, vol. 17, no. 3, pp.400–405, 2002.

[13] M. Niu and Z. Xu, “Efficiency ranking-based evolutionaryalgorithm for power system planning and operation,” IEEETransactions on Power System, vol. 29, no. 3, pp. 1437–1438, 2014.

[14] S. M. Hatefi, F. Jolai, and H. Kor, “A new model for classifyinginputs and outputs and evaluating the DMUs efficiency in DEAbased on Cobb-Douglas production function,” in Proceedingsof the International Conference on Computer Engineering andTechnology (ICCET ’09), pp. 390–394, Singapore, January 2009.

[15] C. Xu and R. Pan, “Network infrastructure and economicgrowth,” in Proceedings of the International Conference onnetworking and Digital Society, Wenzhou, China, 2010.

[16] J. Xu and N. Sun, “The impact of meteorological factors onmanufacturing economy: a quantitative analysis based on trans-log production fuction,” in Proceedings of the InternationalConference on Remote Sensing, Environment and TransportationEngineering (RSETE ’11), pp. 874–877, Nanjing, China, June 2011.

[17] R. Ayres and V. Voudouris, “The economic growth enigma:capital, labour and useful energy?” Energy Policy, vol. 64, pp.16–28, 2014.

[18] M. Filippini, L. C. Hunt, and J. Zoric, “Impact of energypolicy instruments on the estimated level of underlying energyefficiency in the EU residential sector,”Energy Policy, vol. 69, pp.73–81, 2014.

[19] Y. Zhou, S.-C. Huang, and M. Bergsneider, “Linear ridgeregression with spatial constraint for generation of parametricimages in dynamic positron emission tomography studies,”IEEE Transactions on Nuclear Science, vol. 48, no. 1, pp. 125–130,2001.

[20] Shanghai Municipal Electric Power Company, The Report forScience and Technology of Shanghai Power Grids in the TwelfthFive-Year Plan, SMEP, Shanghai, China, 2010.

[21] Z. Hu, B. Shan, and X. Han, The Prospects of Electric PowerDemand in China, CEPP, Beijing, China, 2010.

Submit your manuscripts athttp://www.hindawi.com

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

MathematicsJournal of


Mathematical Problems in Engineering

Hindawi Publishing Corporationhttp://www.hindawi.com

Differential EquationsInternational Journal of

Volume 2014

Applied MathematicsJournal of


Probability and StatisticsHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal of


Mathematical PhysicsAdvances in

Complex AnalysisJournal of


OptimizationJournal of


CombinatoricsHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

International Journal of


Operations ResearchAdvances in

Journal of


Function Spaces

Abstract and Applied AnalysisHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

International Journal of Mathematics and Mathematical Sciences


The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014


Algebra

Discrete Dynamics in Nature and Society



Decision SciencesAdvances in

Discrete MathematicsJournal of

Hindawi Publishing Corporationhttp://www.hindawi.com

Volume 2014 Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Stochastic AnalysisInternational Journal of

Download - Research Article Measuring Technological Progress of Smart ...downloads.hindawi.com/journals/mpe/2014/861820.pdfResearch Article Measuring Technological Progress of Smart Grid Based

Top Related