DOI: 10.3303/CET24114059 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Paper Received: 1 June 2024; Revised: 24 July 2024; Accepted: 28 November 2024 
Please cite this article as: Guo Y., Geng Y., Xu N., Li Z., Liao Q., Liang Y., 2024, Demand Responsive Distributed Energy System Optimization 
Strategy: An Innovative Approach to Oilfield Energy Management  , Chemical Engineering Transactions, 114, 349-354  
DOI:10.3303/CET24114059 
  

 CHEMICAL ENGINEERING TRANSACTIONS  
 

VOL. 114, 2024 

A publication of 

 

The Italian Association 
of Chemical Engineering 
Online at www.cetjournal.it 

Guest Editors: Petar S. Varbanov, Min Zeng, Yee Van Fan, Xuechao Wang 

Copyright © 2024, AIDIC Servizi S.r.l. 

ISBN 979-12-81206-12-0; ISSN 2283-9216 

Demand Responsive Distributed Energy System Optimization 

Strategy: an Innovative Approach to Oilfield Energy 

Management 

Yi Guoa, Yu Genga, Ning Xua, Zhuochao Lia, Qi Liaoa,*, Yongtu Liangb 

aBeijing Key Laboratory of Urban Oil and Gas Distribution Technology, China University of Petroleum-Beijing, Fuxue Road 

No. 18, Changping District, Beijing 102249, China. 
bBeijing University of Chemical Technology, Beijing 100029, China. 

 qliao@cup.edu.cn 

As global demand for sustainable energy grows, so do carbon emissions from the oil and gas sector. This paper 

develops a comprehensive multi-objective mathematical model that incorporates supply-demand dynamics and 

energy storage to optimize both economic and environmental impacts. The model is designed to minimize 

uncertainties in annual operational costs, carbon emissions, and energy demand, providing an advanced 

optimization framework for the sustainable development of oilfield energy systems. Applied to an oilfield in 

Northeast China, the model has reduced annual operational costs and carbon emissions by 16.67 % and 18.36 

%, respectively. This innovative approach based on demand response offers a promising new direction for the 

sustainable management of oilfield energy systems, with significant potential for widespread application and 

practical guidance. 

1. Introduction 

1.1 Background 

As the global demand for sustainable energy continues to grow, the carbon emissions of the oil and gas industry 

also show an upward trend. According to the 2023 Carbon Dioxide Emissions Report released by the 

International Energy Agency, the total global carbon dioxide emissions from energy combustion and industrial 

processes reached 37.4 billion tons in 2023, an increase of 1.1% compared to the previous year (Ruo et al., 

2023). Therefore, the rational adjustment of oil and gas resources and the optimization and upgrading of related 

technologies are extremely urgent. 

In order to effectively reduce energy consumption and carbon emissions in the oil and gas production process, 

distributed energy systems (DES) provide an extremely effective strategy. The system achieves flexibility in 

energy conversion and supply by integrating multiple energy resources (Qin et al., 2021). In addition, DES 

prioritizes the use of clean, low-carbon renewable energy, greatly reducing the cost of energy distribution (Xiao 

et al., 2023). In view of this, energy intensive countries around the world are actively promoting the application 

of DES and conducting extensive research aimed at optimizing the design and operation of DES to achieve 

maximum environmental and economic benefits (Xiao et al., 2023). 

1.2 Related work 

Distributed Energy Systems (DES) can select appropriate energy types based on the resource endowment of 

specific regions, in order to achieve diversified supply of energy services such as electricity, gas, refrigeration, 

and heating. In previous studies, scholars have conducted extensive research, covering multiple aspects such 

as heating networks (Cortés et al., 2018) , power grids (Rn and Sgn, 2023) (Yuan et al., 2024), and heating and 

cooling networks (Rong et al., 2022). Moghimi et al.(2013) identified the optimal design variables and achieved 

multi-objective optimization by defining two objective functions: average annual total cost and efficiency.  

 

349



Compared to independent integrated energy systems, multiple subsystems located in close proximity can 

achieve complementary energy and voltage level support through the interconnection of microgrids. This 

interconnection mode can promote the comprehensive optimization and utilization of distributed energy in the 

region while meeting the operational goals of various subsystems (Rui et al., 2021). Mehleri and his team  

(2012)optimized the design of distributed energy systems (DES) by integrating heating pipeline networks, using 

binary variables to facilitate the connection of pipeline segments. 

The introduction of demand response mechanism aims to solve the problems of energy supply imbalance and 

system stability, and further improve energy utilization efficiency (Dzyuba et al., 2023). Jia et al. (2020) 

considered optimizing the configuration of grid connected optical storage microgrids in response to demand side 

effects. They used an improved particle swarm optimization algorithm to solve the constructed model and 

analyzed the impact of demand side effects on the economic benefits and energy storage configuration of the 

system (Yubin et al., 2020).  

1.3 Research gap and contribution 

Traditional oilfield DES face inefficiencies due to decentralized energy facilities and lack of energy flow. A 

strategy that integrates subsystem interconnection with demand response can optimize DES, balancing load 

peaks, improving energy efficiency, and reducing costs and emissions while enabling users to better manage 

their energy use. The main academic contributions of this paper are as follows: 

• Constructed a multi-objective mathematical programming model that considers supply-demand 

relationships and energy storage mechanisms, aiming to minimize annual costs, carbon emissions, and 

energy demand deviations for a win-win in both economic and environmental benefits. 

• Applies interconnection theory to DES, enabling independent operation and energy sharing with other areas 

through microgrids or pipelines for optimized resource use. 

• Demand response technology is used to adjust energy demand. Through reasonable arrangement of 

adjustable load, energy system efficiency is improved, cost and carbon emissions are reduced, and DES 

operation is optimized. 

 

Figure 1: Technical route 

2. Methodology 

In order to effectively reduce energy consumption and carbon emissions in the oil and gas production process 

and ensure the green and low-carbon operation of the oil field, this chapter constructs a three-objective mixed 

integer linear programming (MILP) model. The model not only considers the two traditional objective functions 

of minimizing annual cost and carbon emissions, but also incorporates the minimization of energy demand 

deviation caused by demand response participation. The model takes energy supply, technology installation 

and operation, and energy balance as constraints to ensure the optimal design and operation scheme. 

350



2.1 Objective function  

Objective function 

Formula (1) shows the calculation model of economic indicators, which divides the annual total cost into three 

parts: total capital cost, total operating cost, and total maintenance cost. 

𝑚𝑖𝑛𝐹1 = 𝑓1 + 𝑓2 + 𝑓3 (1) 

The second side to be addressed is the environmental factor, represented by the total annual carbon emission. 

Eq(2) gives the calculation of carbon emission caused by different energy carrier consumption. 

𝑚𝑖𝑛𝐹2 =∑∑∑∑𝑃𝑑,𝑡,𝑖,𝑒
𝐸𝑁𝐸𝐶𝑐𝑑,𝑡,𝑒

𝐸𝑀𝐼𝑆𝑡𝑑
𝑌𝐸𝐴𝑅𝑡𝑡

𝐷𝐴𝑌

𝑒𝑖𝑡𝑑

 (2) 

The user satisfaction indicator is expressed in terms of total annual energy demand deviation. Maximizing user 

satisfaction means minimizing energy demand deviation. 

𝑚𝑖𝑛𝐹3 =∑∑∑∑|𝑑𝑑,𝑡,𝑖,𝑧
𝑈𝑆𝐸𝑅 − (𝑑𝑑,𝑡,𝑖,𝑧

𝐹𝐼𝑋 + 𝐷𝑑,𝑡,𝑖,𝑧
𝑆𝐻𝐼𝐹𝑇)

𝑧𝑖𝑡𝑑

| +∑∑∑∑|
𝐷𝑑,𝑡,𝑖,𝑧
𝑆𝐻𝐼𝐹𝑇

𝑑𝑑,𝑡,𝑖,𝑧
𝑈𝑆𝐸𝑅

𝑧

−
𝐷𝑑,𝑡−1,𝑖,𝑧
𝑆𝐻𝐼𝐹𝑇

𝑑𝑑,𝑡−1,𝑖,𝑧
𝑈𝑆𝐸𝑅 |

𝑖𝑡𝑑

 (3) 

Energy carrier constraints 

Solar energy is affected by solar irradiance at different times and PV panel size in different subsystems, as 

described in Eq(4). 

𝑃𝑑,𝑡,𝑖,𝑆𝑂𝐿
𝐸𝑁𝐸𝐶 = 𝑆𝑖,𝑃𝑉

𝑇𝐸𝐶𝐻𝑟𝑑,𝑡,𝑖
𝑆𝑂𝐿 (4) 

Take Eq(5) as an example, the input of energy conversion technologies is equal to the total consumption of 

energy carriers.  

𝑃𝑑,𝑡,𝑖,𝑆𝑂𝐿
𝐸𝑁𝐸𝐶 = 𝑃𝑑,𝑡,𝑖,𝑃𝑉

𝑇𝐸𝐼𝑁  (5) 

Technology constraints 

Take the energy conversion technology as an example. As displayed in Eq(6)(7), if an energy conversion 

technology is selected for installation, the size must be lower than its upper bound, upper than its lower bound 

Moreover, the output during operation cannot exceed the size of corresponding energy conversion technology, 

as shown in Eq(8).  

𝑆𝑖,𝑚
𝑇𝐸𝐶𝐻 ≤ 𝐵𝑖,𝑚

𝑇𝐸𝐼𝑁𝑈𝐵𝑖,𝑚 (6) 

𝐵𝑖,𝑚
𝑇𝐸𝐼𝑁𝐿𝐵𝑖,𝑚 ≤ 𝑆𝑖,𝑚

𝑇𝐸𝐶𝐻 (7) 

𝑃𝑑,𝑡,𝑖,𝑚
𝑇𝑂𝑈𝑇 ≤ 𝑆𝑖,𝑚

𝑇𝐸𝐶𝐻 (8) 

According to Eq(9), if an energy conversion technique is not selected for operation, the corresponding output is 

equal to zero. However, if an energy conversion technique is selected for operation, the corresponding output 

must be upper than the minimum value, as shown in equation (10). 

𝑃𝑑,𝑡,𝑖,𝑚
𝑇𝑂𝑈𝑇 ≤ 𝑀1𝐵𝑑,𝑡,𝑖,𝑚

𝑇𝐸𝑂𝑃  (9) 

𝑆𝑖,𝑚
𝑇𝐸𝐶𝐻𝛾𝑚 ≤ 𝑃𝑑,𝑡,𝑖,𝑚

𝑇𝑂𝑈𝑇 +𝑀1(1 − 𝐵𝑑,𝑡,𝑖,𝑚
𝑇𝐸𝑂𝑃 ) (10) 

Energy balance constraints 

The adjusted energy demand is equal to the sum of the fixed energy demand and the removable energy 

demand. It must meet a rational upper and lower bound and the total energy demand keeps constant in a day, 

seen in Eq(11) and Eq(12). 

𝑑𝑑,𝑡,𝑖,𝑗
𝑀𝐼𝑁 ≤ 𝑑𝑑,𝑡,𝑖,𝑗

𝐹𝐼𝑋 + 𝐷𝑑,𝑡,𝑖,𝑗
𝑆𝐻𝐼𝐹𝑇 ≤ 𝑑𝑑,𝑡,𝑖,𝑗

𝑀𝐴𝑋  (11) 

∑𝐷𝑑,𝑡,𝑖,𝑗
𝑆𝐻𝐼𝐹𝑇

𝑡

+∑𝑑𝑑,𝑡,𝑖,𝑗
𝐹𝐼𝑋

𝑡

=∑𝑑𝑑,𝑡,𝑖,𝑗
𝑈𝑆𝐸𝑅

𝑡

 (12) 

351



Take the power balance for example, the electricity balance is given in Eq(13).  

𝑃𝑑,𝑡,𝑖,𝑃𝑉
𝑈𝑆𝐸𝑅 + 𝑃𝑑,𝑡,𝑖,𝐺𝑇

𝑈𝑆𝐸𝑅 + 𝐸𝑑,𝑡,𝑖
𝑀𝐺 + 𝑃𝑑,𝑡,𝑖,𝐺𝑅

𝐸𝑁𝐸𝐶 + 𝑃𝑑,𝑡,𝑖,𝐸𝐿𝐸𝐶
𝑆𝑂𝑈𝑇 − 𝑃𝑑,𝑡,𝑖,𝐸𝐿𝐸𝐶

𝑆𝑇𝐼𝑁 − 𝑃𝑑,𝑡,𝑖,𝐸𝑅
𝑇𝐸𝐼𝑁 = 𝑑𝑑,𝑡,𝑖,𝐸𝐿𝐸𝐶

𝐹𝐼𝑋 + 𝐷𝑑,𝑡,𝑖,𝐸𝐿𝐸𝐶
𝑆𝐻𝐼𝐹𝑇  (13) 

3. Cases and Results 

3.1 Background description 

The model is universal and can be set according to specific conditions. This study takes an oil field in Northeast 

China as an example to verify the effectiveness of the model. The oilfield uses electric heating technology to 

maintain the appropriate process temperature of the medium and prevent the solidification of crude oil pipelines. 

It consists of 30 wells, 5 valve groups, 1 oil station and 1 combined treatment station. It is divided into 7 

independent working areas, each of which includes a gas station and its affiliated oil wells. 

There are significant differences in energy demand across different seasons throughout the year. For example, 

in winter, due to low temperatures, in order to prevent crude oil from solidifying, it is necessary to increase the 

heating temperature, which leads to a sharp increase in the demand for electricity in electric heating pipelines; 

In summer, electricity demand is relatively reduced. The joint station where W7 is located is responsible for oil, 

gas, and water treatment, with a variety of equipment, resulting in a higher demand for multiple energy sources. 

The W1-W5 work area is mainly composed of valve groups and their managed wellhead, with a relatively single 

work nature and no steam demand throughout the year. Considering the low winter temperatures in the area 

where the oil field is located, the demand for electricity, cooling, heating, and steam in each work area fluctuates 

significantly with seasonal and temperature changes. 

On the basis of comprehensive consideration of various factors, this study selects the carbon emission 

coefficient of electricity as 890 g CO2 / kWh and natural gas as 900 g CO2 / kWh. Since solar power generation 

does not involve the combustion process, its carbon emission coefficient is determined to be 0, highlighting the 

environmental advantages of solar energy as a zero-emission renewable energy source. According to the actual 

working ability and infrastructure of different working areas, the selected equipment installation capabilities of 

the seven areas are different. Table 1 summarizes the working hours, efficiency, and cost of various types of 

equipment: 

Table 1 Equipment running time and efficiency 

Device Name 
Device life 

(a) 

operating efficiency 

(%) 

 Fixed capital cost 

(CNY) 
 Linear capital cost 

PV 30 17 250000 2,800 CNY/m² 

GT 30 33 100000 1,500 CNY/kW 

GB 20 83 20000 850 CNY/kW 

BB 25 85 50000 600 CNY/kW 

ER 25 430 150000 1,640 CNY/kW 

SC 25 133 100000 970 CNY/kW 

HE 20 98 70000 210 CNY/kW 

WB 15 78 12000 1,200 CNY/kW 

ES 15 90/85 150000 950 CNY/kWh 

HS 20 90/85 80000 90 CNY/kWh 

CS 20 90/85 80000 190 CNY/kWh 

3.2 Running results 

GUROBI 10.0.2 is used to solve the model.In the single objective function analysis, the annual cost ( F1 ) is 

8.12 × 106 CNY, the annual carbon emission ( F2 ) is 1.37 × 103 t, and the annual energy demand deviation ( 

F3 ) is 2.97 × 103 kWh. In order to ensure environmental friendliness and energy supply stability, the ε-constraint 

method is used to balance the three objectives, with F1 as the optimization objective and F2 and F3 as 

constraints. F1 and F2 are divided into different intervals, forming a multi-combination parameter set. The 

optimization results shown in Figure 2 show that all 20 schemes achieve Pareto optimality. Among all schemes, 

the ninth scheme achieves an equilibrium among the three objectives. Although it is not optimal for each 

objective, it is considered to be the most consistent choice with the preferences of current decision makers. 

Figure 2 shows the final linearization metrics for the three objectives in the 20 scenarios, where the red bar 

represents F1, purple represents F2, and orange represents F3. Figure 3 shows the relationship between annual 

cost and energy demand deviation, which provides a reference for decision makers to formulate more 

352



comprehensive pricing measures and incentives. In the ninth scheme, the energy demand deviation can be 

reduced by 168.71 × 103 kWh. 

 

Figure 2: Linearization indexes of different targets in 20 schemes. 

  

Figure3: The cost and energy demand deviation in Scheme 9. 

After optimization, the capacity of each equipment is reduced, reflecting the original excess energy configuration 

of the oilfield, indicating that there is additional energy that can be stored. The difference of installed capacity 

before and after optimization is shown in Table 2, and the details of energy storage state are shown in Table 3. 

Table 2 The difference of installed capacity before and after optimization 

Device 

Name 

PV GT GB BB ER SC HE WB ES HS CS 

W1 0 900 269.39 0 833.68 357.59 0 900 1500 1500 645.03 

W2 0 900 534.7 0 870 681.36 260 900 1500 1500 565.45 

W3 0 900 875.61 0 839.22 864 594.1 900 1500 1500 1430.96 

W4 0 900 696.28 0 869.39 778.08 418.36 900 1500 1500 1186.52 

W5 0 6000 147.16 0 5510.38 4600 2264.21 6000 0 0 6651.74 

W6 0 6000 2471.94 0 5100 3800 3072.5 6000 444.23 2607.97 8000 

W7 0 6000 1033.445 0 5648 5179.02 2525.18 6000 1138.82 3482.36 8000 

Table 3 Energy storage situation 

Item W1 W2 W3 W4 W5 W6 W7 

Electric storage 0 0 0 0 385.3001 300.7242 285.6198 

Thermal storage 0 0 0 0 222.0138 101.964 43.373 

Cold storage 20495.47 15363.79 2064.36 4829.037 49.5862 0 0 

353



The optimization scheme performs better in economy, which is 16.67 % lower than the original scheme, and the 

carbon emission is reduced by 18.36 %, which significantly promotes the green and low-carbon operation of the 

oilfield. The specific comparison details are shown in Table 4. 

Table 4 Economic situation comparison 

Comparative item 
Installation costs 

（×106CNY) 

Operating costs 

（×106CNY) 

Maintenance costs 

（×106CNY) 

Carbon emissions 

（×103t/year) 

Original plan 4.02 3.37 1.86 1.6 

Optimized plan 3.60 2.80 1.60 1.35 

Reduce the 

proportion（%） 
11.7 20.5 16.3 15.52 

4. Conclusion 

This research contributes a novel approach for optimizing sustainable energy management in oilfields, 

leveraging the synergy of demand response and distributed energy systems. A rigorous multi-objective 

mathematical programming model is formulated to concurrently minimize annual costs, carbon emissions, and 

energy demand deviation. The model encapsulates constraints pertaining to energy supply, technology, 

interconnection, and demand response, thus enabling the system to flexibly adapt energy management and 

optimize the design and operation of distributed energy systems. The application of the model to an oilfield case 

study in Northeast China resulted in a remarkable18.36 % reduction in annual operational costs and a 16.67% 

decrease in carbon emissions. This innovative optimization framework, underpinned by demand response, 

promises significant potential for enhancing sustainable energy management practices in oilfields, showcasing 

its broad applicability and substantial practical significance in the field. 

Reference 

Ameri M, Besharati Z., 2016, Optimal design and operation of district heating and cooling networks with CCHP 

systems in a residential complex [J]. Energy & Buildings, 110: 135-48. 

Cortés P, Muñuzuri J, BERROCAL-DE-O M, 2018, Genetic algorithms to optimize the operating costs of 

electricity and heating networks in buildings considering distributed energy generation and storage [J]. 

Computers and Operations Research, 96: 157-72. 

Dzyuba A, Solovyeva I, Semikolenov A., 2023, Raising the Resilience of Industrial Manufacturers through 

Implementing Natural Gas-Fired Distributed Energy Resource Systems with Demand Response [J]. 

Sustainability, 15(10). 

Mago P J, Hueffed A K., 2010, Evaluation of a turbine driven CCHP system for large office buildings under 

different operating strategies [J]. Energy & Buildings, 42(10): 1628-36. 

Mehleri E D, Sarimveis H, Markatos N C, 2012, A mathematical programming approach for optimal design of 

distributed energy systems at the neighbourhood level [J]. Energy, 44(1): 96-104. 

Moghimi H, Ahmadi A, Aghaei J, 2013, Stochastic techno-economic operation of power systems in the presence 

of distributed energy resources [J]. International Journal of Electrical Power and Energy Systems, 45(1): 

477-88. 

Qin Z, Haoran Z, Yamin Y, 2021, Sustainable and clean oilfield development: How access to wind power can 

make offshore platforms more sustainable with production stability [J]. Journal of Cleaner Production, 294.  

Rong L C, Fang T Y, Qing W H, 2022, Comparison of life cycle performance of distributed energy system and 

conventional energy system for district heating and cooling in China [J]. Journal of Central South University, 

29(7): 2357-76. 

R N, S G N., 2023, Optimal Power Flow Management in Microgrids using Distributed Energy Resources [J]. IOP 

Conference Series: Materials Science and Engineering, 1295(1). 

Rui Q, Qi L, Jie Y, 2021, The coupling impact of subsystem interconnection and demand response on the 

distributed energy systems: A case study of the composite community in China [J]. Energy, 228. 

Ruochong X, Dan T, Qingyang X, 2023, MEIC-global-CO2: A new global CO2 emission inventory with highly-

resolved source category and sub-country information [J]. Science China Earth Sciences, 67(2): 450-65.  

Xiaoyu Z, Xingxing Z, Pu G, 2023, A review of distributed energy system optimization for building 

decarbonization [J]. Journal of Building Engineering, 73.  

Yuan Y, Yuan X, Wang H, 2024, Model exploration of grid adjustment and restoration strategy based on 

intelligent decision system [J]. International Journal of Thermofluids, 21: 100580-. 

Yubin J, Yang D Z, Changyin S, 2020, Distributed Economic Model Predictive Control for a Wind–Photovoltaic–

Battery Microgrid Power System [J]. IEEE Transactions on Sustainable Energy, 11(2): 1089-99.  

354